Stephen P. Osborne Editor the New Public Governance Emerging Perspectives on the Theory and Practice...

THE NEW PUBLIC GOVERNANCE?

‘Stephen Osborne is to be congratulated for assembling an impressive arrayof contributors and producing a text which makes a major contribution to the debates about this emerging paradigm.’ Robert Pyper GlasgowCaledonian University, UK

‘This book showcases the best writing on “public governance”, bring-ing together new research in institutional analysis, policy dynamics, and government–society relationships. It provides a good balance of conceptualinsights and empirical analysis grounded in the large changes evident inadvanced countries in recent decades.’ Brian Head University of Queensland,Australia

‘This text, edited by Stephen Osborne, may do as much for critical insightto Europe’s public service delivery as the David Osborne and Ted Gaeblertext on Reinventing Government did for the United States in the 1990s in refocusing the public sector for improvement.’ Dean F. Eitel DePaulUniversity, USA

Despite predictions that ‘new public management’ would establish itself as thenew paradigm of Public Administration and Management, recent academicresearch has highlighted concerns about the intra-organizational focus and limitations of this approach. This book represents a comprehensive analysis ofthe state of the art of public management, examining and framing the debate inthis important area.

The New Public Governance? sets out to explore this emergent field of researchand to present a framework with which to understand it. Divided into five parts,it examines:

• Theoretical underpinnings of the concept of governance, especially compet-ing perspectives from Europe and the US

• Governance of inter-organizational partnerships and contractual relationships• Governance of policy networks• Lessons learned and future directions

Under the steely editorship of Stephen Osborne, and with contributions fromleading academics including Owen Hughes, John M. Bryson, Don Kettl, GuyPeters and Carsten Greve, this book will be of particular interest to researchersand students of public administration, public management, public policy and public services management.

Stephen P. Osborne is Professor of International Public Management and Directorof the Centre for Public Services Research at the University of Edinburgh, UK.He is President of the International Research Society for Public Management(IRSPM) and Editor of Public Management Review (PMR).

First published 2010

by Routledge2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN

Simultaneously published in the USA and Canadaby Routledge

270 Madison Ave, New York, NY 10016

Routledge is an imprint of the Taylor & Francis Group, an informa business

or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or

retrieval system, without permission in writing from the publishers.

British Library Cataloguing in Publication DataA catalogue record for this book is available

from the British Library

Library of Congress Cataloging in Publication DataThe new public governance? : emerging perspectives on the theoryand practice of public governance / edited by Stephen P. Osborne.

p. cm.Includes bibliographical references and index.

1. Public administration. 2. Public-private sector cooperation.3. Organizational change. I. Osborne, Stephen P., 1953–

JF1351.N44 2009352.01–dc222009027964

ISBN10: 0-415-49462-1 (hbk)ISBN10: 0-415-49463-X (pbk)ISBN10: 0-203-86168-X (ebk)

ISBN13: 978-0-415-49462-5 (hbk)ISBN13: 978-0-415-49463-2 (pbk)ISBN13: 978-0-203-86168-4 (ebk)

This edition published in the Taylor & Francis e-Library, 2010.

To purchase your own copy of this or any of Taylor & Francis or Routledge’scollection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.

ISBN 0-203-86168-X Master e-book ISBN

IN LOVING MEMORY OF MY FATHER, FRANK OSBORNE

CONTENTS

List of tables xList of figures xiContributors xiii

1 Introduction 1The (New) Public Governance: a suitable case for treatment?STEPHEN P. OSBORNE

PART ITheoretical perspectives on public governance 17

2 Global perspectives on governance 19PATRICIA KENNETT

3 Meta-governance and public management 36B. GUY PETERS

4 Innovations in governance 52MARK MOORE AND JEAN HARTLEY

5 Governance and governability 72JAN KOOIMAN

6 Does governance exist? 87OWEN HUGHES

7 What endures? Public governance and the cycle of reform 105LAURENCE E. LYNN, JR

PART IIGovernance and interorganizational partnerships to deliver public services 125

8 Theory of organizational partnerships: partnership advantages,disadvantages and success factors 127RONALD W. MCQUAID

9 Public–private partnerships and public governance challenges 149CARSTEN GREVE AND GRAEME HODGE

10 Introducing the theory of collaborative advantage 163SIV VANGEN AND CHRIS HUXHAM

11 Relationship marketing, relational capital and the governance of public services delivery 185STEPHEN P. OSBORNE, KATE MCLAUGHLIN AND

CELINE CHEW

12 Leading across frontiers: how visionary leaders integrate people, processes, structures and resources 200BARBARA C. CROSBY, JOHN M. BRYSON AND

MELISSA M. STONE

13 Public governance and the third sector: opportunities for co-production and innovation? 223VICTOR PESTOFF AND TACO BRANDSEN

PART IIIGovernance of contractual relationships 237

14 Governance, contract management and public management 239DONALD F. KETTL

15 Governance of outsourcing and contractual relationships 255FEDERICA FARNETI, EMANUELE PADOVANI AND

DAVID W. YOUNG

16 The governance of contracting relationships: “killing the golden goose” 270A third-sector perspectiveSTEVEN RATHGEB SMITH AND JUDITH SMYTH

CONTENTS

PART IVGovernance of interorganizational networks 301

17 Trust in governance networks: looking for conditions for innovative solutions and outcomes 303ERIK-HANS KLIJN

18 Implementation and managerial networking in the New PublicGovernance 322LAURENCE J. O’TOOLE, JR, AND KENNETH J. MEIER

19 From new public management to networked community governance? Strategic local public service networks in England 337STEVE MARTIN

PART VGovernance of policy networks 349

20 Policy networks: theory and practice 351TOBIAS JUNG

21 Policy networks in practice: the debate on the future of Amsterdam Airport Schiphol 365MENNO HUYS AND JOOP KOPPENJAN

22 Governance, networks and policy change: the case of cannabis in the United Kingdom 394BEATRIZ ACEVEDO AND RICHARD COMMON

23 Conclusions 413Public governance and public services delivery: a research agenda for the futureSTEPHEN P. OSBORNE

Index 429

CONTENTS

TABLES

1.1 Core elements of the NPG, in contrast to PA and the NPM 10

11.1 Commissioner–provider inter-relationships in public services provision (developed and adapted from the work of Ring and Van de Ven 1992: 490) 193

14.1 Presence of contractor personnel during US military operations 241

16.1 Impact of performance contracting on governance 28316.2 Impact of external regulation on governance 28816.3 Impact of competition on governance 29222.1 Classification of illegal drugs in the United Kingdom, based

on the Misuse of Drugs Act, 1971 40022.2 Policy networks involved in the policy-making on cannabis

in the United Kingdom 40423.1 An exploratory model of the interaction of public policy

implementation and public services delivery regimes and managerial practice 414

FIGURES

2.1 Who gains? 302.2 Shift in voting power toward developing countries 30

10.1 Themes in collaboration practice 16410.2 Aims in collaboration practice 16510.3 The trust-building loop 16810.4 Tensions in the management of culture 17510.5 Examples of stances to giving knowledge (reprinted with

permission from Blackwell Publishing from Huxham and Hibbert 2008) 179

10.6 Examples of stances to taking knowledge (reprinted with permission from Blackwell Publishing from Huxham and Hibbert 2008) 180

12.1 The triple three-dimensional view of power 20212.2 A framework for understanding leadership in cross-sector

collaborations 20415.1 Public–private partnerships: the service delivery

spectrum 25715.2 The four governance models 25715.3 A framework for assessing risk 26215.4 Examples of outsourcing risk 26315.5 Outsourcing risk and the Four-governance Model 26421.1 Spatial situation Schiphol region 2006 37021.2 The new Schiphol policy arena (at the centre) with actors

from three different networks 37321.3 Schiphol policy arena anno 1997 with Schiphol

growth coalition (left) and environmental coalition (right) 377

21.4 Mismatch between calculated, legally embedded noise contours (dots) and actual noise pollution (grey area) around Schiphol, 2003–6 380

21.5 Schiphol policy arena anno 2006 with Schiphol growth coalition (left) and environmental coalition (right) 380

21.6 Schiphol policy arena anno 2008 with Schiphol growth coalition (left) and environmental coalition (right) 383

22.1 International agencies, governmental institutions and diverse actors involved in the debate about cannabis policy in the United Kingdom 402

FIGURES

CONTRIBUTORS

Beatriz Acevedo is a researcher on international drugs policy at AngliaRuskin University.

Taco Brandsen is Associate Professor at Radboud University Nijmegen.

John M. Bryson is McKnight Presidential Professor of Planning and PublicAffairs at the University of Minnesota.

Celine Chew is Lecturer in Marketing and Strategy at Cardiff University.

Richard Common is a Senior Research Fellow at the .

Barbara C. Crosby is Associate Professor in Public and Non-profit Leader-ship at the University of Minnesota.

Federica Farneti is Assistant Professor at the Alma Mater StudiorumUniversity of Bologna.

Carsten Greve is Professor of Public Management at Copenhagen BusinessSchool.

Jean Hartley is Professor of Organizational Analysis at the Institute ofGovernance and Public Management, University of Warwick.

Graeme Hodge is a Professor of Law at Monash University, and Directorof the Monash Center for Regulatory Studies.

Owen Hughes is Professor of Public Management at Monash University.

Chris Huxham is Professor of Management at the University of StrathclydeBusiness School.

Menno Huys is a PhD student at the Faculty of Policy and Management atDelft University of Technology.

Tobias Jung is ESRC Research Fellow within the Centre for Charitable Givingand Philanthropy, and based at the University of Edinburgh Business School.

Patricia Kennett is Reader in Comparative Policy Studies at the Universityof Bristol.

Donald F. Kettl is Dean of the School of Public Policy at the University ofMaryland.

Erik-Hans Klijn is Professor of Public Administration at Erasmus University.

Jan Kooiman is now retired. He was Professor of Public Organisations atErasmus University.

Joop Koppenjan is Associate Professor of Public Management at the DelftUniversity of Technology.

Laurence E. Lynn, Jr, is the Sydney Stein, Jr, Professor of Public Manage-ment at the University of Chicago.

Kate McLaughlin is now retired. She was Senior Lecturer in Local Govern-ance at the University of Birmingham.

Ronald W. McQuaid is Director of the Employment Research Institute atEdinburgh Napier University.

Steve Martin is Professor of Public Management at Cardiff University.

Kenneth J. Meier is the Distinguished Professor of Political Science at TexasA. & M. University, and Professor of Public Management at CardiffUniversity.

Mark Moore is Hauser Professor of Non-profit Organizations, J. F.Kennedy School of Government, at Harvard University.

Stephen P. Osborne is Professor of International Public Management at theUniversity of Edinburgh Business School.

Laurence J. O’Toole, Jr, is the Margaret Hughes and Robert T. GolembiewskiProfessor of Public Administration at the University of Georgia.

Emanuele Padovani is Associate Professor at the Alma Mater StudioromUniversity of Bologna.

Victor Pestoff is Guest Professor both at the Institute for Civil Society Studies, Ersta Sköndal University College, Stockholm, and at the Euro-pean Research Institute on Cooperative and Social Enterprise, Universityof Trento.

B. Guy Peters is Maurice Falk Professor of Government at the Universityof Pittsburgh.

Steven Rathgeb Smith is the Nancy Bell Evans Professor of Public Affairsat the University of Washington.

Judith Smyth is director for public governance and commissioning with OPM,England.

CONTRIBUTORS

Melissa M. Stone is a member of the Humphrey Institute of Public Affairs,University of Minnesota.

Siv Vangen is a senior lecturer at the Open University, England.

David W. Young is Emeritus Professor of Management at Boston University.

CONTRIBUTORS

INTRODUCTION

The (New) Public Governance: a suitable case for treatment?1

Stephen P. Osborne

Introduction

More than a decade has passed since the publication of Christopher Hood’sinfluential piece that codified the nature of the New Public Management (NPM)paradigm (Hood 1991). At that time it seemed likely, certainly within theAnglo-American research community, that this paradigm would sweep all before it in its triumphal recasting of the nature of our discipline – in theory and in practice. A hundred-odd years of the hegemony of PublicAdministration (PA) in the public sphere seemingly counted for nothing inthis momentous shift. Since then, though, the debate on the impact of theNPM upon the discipline, and indeed about whether it is a paradigm at all(Gow and Dufour 2000), has become more contested.2

This introductory chapter is intended to move this debate forward. It con-siders, somewhat provocatively, that the NPM has actually been a transitorystage in the evolution from traditional Public Administration to what is herecalled the New Public Governance.3 A note upon terminology is importanthere. The term “public policy implementation and public services delivery”is used here to denote the overall field of the design and implementation of public policy and the delivery of public services. Within this, PublicAdministration, the NPM and the New Public Governance (NPG) are thendenoted as policy and implementation regimes within this overall field – thusneatly skirting the above, rather redundant, argument as to whether theseregimes are actually paradigms or not.

The argument advanced here is that public policy implementation and public services delivery have passed through three design and deliveryregimes: a longer, pre-eminent one of PA, from the late nineteenth centurythrough to the late 1970s/early 1980s; a second one, of the NPM, throughto the start of the twenty-first century; and an emergent third one, of the

NPG, since then. The time of the NPM has thus in fact been a relativelyshort-lived and transient one between the statist and bureaucratic traditionof PA and the embryonic plural and pluralist tradition of the NPG. Theremainder of this chapter will first expound upon the extant natures of PAand the NPM. It will then explore the nature of public governance and the NPG before considering the new challenges that it poses for both thetheory and the practice of public policy implementation and public servicesdelivery.

Inevitably, such a tripartite regime model is a simplification – elements of each regime can and will coexist with each other or overlap. Many network governance systems often operate in the shadow of, or in spite of, the dominant regime of hierarchy – for example, both PA and public governance contain strong, if differentiated, elements of hierarchy (Klijn 2002). The intention here is to tease out three “archetypes”, in the Weberiantradition, that will assist and promote analysis and discussion of the conceptual and practical development of public policy implementation andpublic services delivery.

It must be emphasized that this book is not meant to propose “the NPG” as a new paradigm of public services delivery. It is neither that normative nor that prescriptive. The question mark in the title is deliberate. Rather, this book is a critical examination of the concept of “public governance”.Offering a range of perspectives, the book questions whether or not publicgovernance is a new paradigm for the delivery of public services in the twenty-first century, and offers a range of critical perspectives upon it – bothin theory and in practice.

In entering into this discussion, it is useful to bear in mind a distinctionmade by Dawson and Dargie (1999) in their work on the NPM. They urgethe necessity to differentiate between the NPM as a political ideology, as an academic field of study and as a body of managerial practice. The lattertwo elements are of especial importance and need to be held distinct. The analysis of the workings and impact of any regime is distinct from thenormative assertion of “how best” to manage within it.

The shadow of the past . . .

Public Administration

The key elements of PA4 (Hood 1991) can be defined as

• The dominance of the “rule of law”;• A focus on administering set rules and guidelines;• A central role for the bureaucracy in making and implementing policy;• The “politics–administration” split within public organizations;

INTRODUCTION

• A commitment to incremental budgeting; and• The hegemony of the professional in public service delivery.

Developing out of the early years of the public sector in the late nineteenthand early twentieth centuries, PA, as an academic field of study, has been astrongly “grounded”, rather than theoretical, discipline in the UK – the classicearly statement being Robson 1928 – and at variance with its cousins in main-land Europe and the US, which are more firmly located within administrativetheory.

As a field of practice, it reached its high point in the UK in the 1945–79 eraof the welfare state, when the state was confidently expected to meet all thesocial and economic needs of the citizenry, “from the cradle to the grave”.PA was to be the instrument of this brave new world, with a focus on adminis-trative procedures to ensure equality of treatment. Predictably, perhaps, such a vision was doomed to failure – public needs inevitably outstrippedthe public resources available to meet them. In the latter days of their hegem-ony both the welfare state and PA came under increasing fire – first fromtheir academic critics (for example, Dunleavy 1985) and eventually from the political elite (see Mischra 1984 for an overview of these critiques). Mostdamagingly, Chandler (1991) argued that PA had now entered terminal decline as a discipline, whilst Rhodes (1997) asserted that it had become a “bystander” to the practice of public policy implementation and public services delivery. This paved the way for the rise of the NPM.

New Public Management (NPM)

The spread of the NPM, from the late 1970s onward, saw the growth of anew discourse of public policy implementation and public services delivery.In its most extreme form, this asserted the superiority of private-sector man-agerial techniques over those of PA, with the assumption that the applicationof such techniques to public services delivery would automatically lead toimprovements in the efficiency and effectiveness of these services (Thatcher1995). The key elements of the NPM can be summarized as:

• An attention to lessons from private-sector management;• The growth both of hands-on “management” – in its own right and not

as an offshoot of professionalism – and of “arm’s length” organizationswhere policy implementation was organizationally distanced from the policy-makers (as opposed to the “interpersonal” distancing of the policy–administration split within PA);

• A focus upon entrepreneurial leadership within public service organizations;• An emphasis on inputs and output control and evaluation, and upon

performance management and audit;

INTRODUCTION

• The disaggregation of public services to their most basic units and a focuson their cost management; and

• Within the Anglo-American and Australia/New Zealand regions at least,the growth of use of markets, competition and contracts for resourceallocation and service delivery within public services.

In the research community, this led to a focus upon the management of public services and of public service organizations (PSOs) as a distinct fieldseparate from the public policy process – public management as opposed to public administration. At a practical level, it led to the evolution of management as a coherent and legitimized role and function within PSOs, incontrast to (and often in conflict with) the traditional professional groupingswithin PSOs.

In the years since it first contested the territory of public policy imple-mentation and public services delivery with PA, though, the nature and/orsuccess(es) of the NPM have been questioned on a range of grounds (seeMcLaughlin et al. 2002 for an overview of these critiques). Critics have arguedinter alia that:

• The NPM is not one phenomenon or paradigm, but a cluster of several(Ferlie et al. 1996) – and has a number of distinct personae, dependentupon the audience, including ideological, managerial and research-oriented personae, as discussed by Dawson and Dargie (1999) above;

• The geographic extent of the NPM is limited to the Anglo-American,Australasian and (some) Scandinavian arenas, whilst PA continues toremain dominant elsewhere (Kickert 1997; see also Hood 1995);

• The nature of the NPM itself is also geographically variegated – with,for example, the British and American variants actually being quite distinct from each other in their focus and locus (Borins 2002);

• In reality, the NPM is simply a subschool of PA that has been limitedin its impact by the lack of a real theoretical base and conceptual rigor(Frederickson and Smith 2003);

• The benefits of the NPM are at best partial and contested (Pollitt andBouckaert 2004); and

• That the NPM was a “disaster waiting to happen” (Hood and Jackson1992) and was a failed paradigm (Farnham and Horton 1996).

Similarly, in the dedicated textbooks on this topic, one will find both advo-cates of the NPM (Hughes 2002) and critics (Flynn 2002).

The NPM has been criticized most devastatingly for its intraorganizationalfocus in an increasingly plural world and for its adherence to the applicationof outdated private-sector techniques to public policy implementation andpublic services delivery – and in the face of evidence about their inapplicabil-ity (Metcalfe and Richards 1991).

INTRODUCTION

The state of the art

Increasingly, then, both PA and the NPM have begun to look like partialtheories, at best. The strength of PA is in its exploration of the essentiallypolitical nature of public policy implementation and public services delivery,and of the complexities and nuances of the public policy making and imple-mentation process. However, the extent to which the implementation studiesliterature within PA has been able to unpackage the differential influencesupon public policy implementation has been disappointing. There is a tendencyfor implementation to be seen simply as a “black box” with no apparent will to explore the complex subprocesses of the management of the outputsof the policy process – public services themselves (Schofield 2001; Hill andHupe 2003). At worst, public managers and management are portrayed asthe villain(s) of the piece, thwarting the resolve of their political masters andoften subverting the intentions of new policy to their own ends.

By contrast, the strength of the NPM has been in its ability to addressprecisely the complexities of this “black box”, now recodified as the man-agement of change and innovation (Osborne and Brown 2006). However, it has suffered with an equally limiting tendency to see the public policy process as simply a “context” within which the essential task of public man-agement takes place. In its most extreme form, the NPM has even questionedthe legitimacy of public policy as a context for public management, arguingthat it imposes unreasonable democratic constraints on the management and provision of public services (Meier 1997). Most damagingly, though, isthe argument that the NPM is limited and one-dimensional in its ability to capture and contribute to the management and governance of public services and of PSOs in an increasingly fragmented and interorganizationalenvironment (Rhodes 1997).

The argument made here is hence that both PA and the NPM fail to cap-ture the complex reality of the design, delivery and management of publicservices in the twenty-first century. Given such criticisms of both paradigms,therefore, it is time to question whether there is a pressing need now for a more sophisticated understanding of public policy implementation and public services delivery – one that moves beyond the sterile dichotomy of“administration versus management” and that allows a more comprehen-sive and integrated approach to the study, and practice, of public policy implementation and public services delivery. The intention in this volume is to explore whether the NPG has the potential or actuality to be this over-arching theoretical framework for the study and the practice of public services delivery.

INTRODUCTION

. . . And the shadow of the future

The New Public Governance (NPG)

At the outset, it is important to be clear, once more, about two points. First,that the NPG is being presented here neither as a normative new paradigmto supersede PA and the NPM nor as “the one best way” (Alford and Hughes2008) to respond to the challenges of public policy implementation and public services delivery in the twenty-first century. Rather it is being presentedboth as a conceptual tool with the potential to assist our understanding of the complexity of these challenges and as a reflection of the reality of theworking lives of public managers today.

Second, “governance” and “public governance” are not new terms – they come with considerable prior theoretical and/or ideological baggage. Critics have differentiated three broad schools of governance literature: corporate governance, “good” governance, and public governance.

Corporate governance is concerned with the internal systems and processesthat provide direction and accountability to any organization. In public services it has most often been concerned with the relationship between thepolicy-makers and/or trustees of public organizations and the senior man-agers given the task of making these policies a reality (for example, Cornforth2003).

“Good” governance is concerned with the promulgation of normativemodels of social, political and administrative governance by supranationalbodies such as the World Bank (Leftwich 1993; Rhodes 1997). Invariablythis has placed a premium upon market-based approaches to the alloca-tion and governance of public resources (see, for example, Osborne andKaposvari 1997).

Public governance, which is the focus here, can itself be broken down intofive distinct strands:

• Socio-political governance, concerned with the over-arching institutionalrelationships within society. Kooiman (1999) argues that these relation-ships and interactions must be understood in their totality in order tounderstand the creation and implementation of public policy. In thisapproach, government is no longer pre-eminent in public policy but has to rely upon other societal actors for its legitimacy and impact inthis field.

• Public policy governance, concerned with how policy elites and networksinteract to create and govern the public policy process. Marsh andRhodes (1992), Börzel (1997) and Klijn and Koppenjan (2000), build-ing upon the work of Hanf and Scharpf (1978), are good examples ofsuch explorations of the workings of policy communities and networks.Most recently, Peters (2008) has explored “meta-governance” instruments

INTRODUCTION

as a way by which to reassert political direction within multi-stakeholderpolicy networks.

• Administrative governance, concerned with the effective application of PA and its repositioning to encompass the complexities of the contem-porary state. Thus, for example, Salamon (2002) uses governance almost as a proxy term for the generic practice of public policy implementa-tion and public services delivery, whilst Lynn et al. (2001) also use it as a catch-all term to try to create a holistic theory of public policy implementation and public services delivery in conditions of the “hollowstate” (Milward and Provan 2003). More provocatively, Frederickson(1999) contends that governance, taken together with the theory of“administrative conjunction”, is in fact a way to reposition PA as thecontinuing pre-eminent discipline for the realities of the modern world.

• Contract governance, concerned with the inner workings of the NPM,and particularly the governance of contractual relationships in the delivery of public services. In this vein, Kettl has argued that public agencies in the modern contract state have become “responsible for a[public service delivery] system over which they [have] little control” (Kettl1993: 207; see also Kettl 2000).

• Network governance, concerned with how “self organizing inter-organizational networks” (Rhodes 1997; see also Kickert 1993) functionboth with and without government to provide public services. In con-trast to public policy governance, this is focused upon those networksthat implement public policy and deliver public services (for exampleDenters and Rose 2005; Entwistle and Martin 2005).

All of these theoretical perspectives on governance make an important contribution to our understanding of public policy implementation andpublic services delivery. The intention here is to argue that, from being anelement within the PA and NPM regimes of public policy implementationand public services delivery, public governance has become a distinctive regime in its own right – the NPG. The intention here is to suggest and explorea distinctive niche for the NPG that captures the realities of public policyimplementation and public services delivery within the plural and pluralistcomplexities of the state in the twenty-first century.

Working with the above definitions of public governance, therefore, andbuilding upon the insights of Peters and Pierre (1998), it is argued here thatit is possible, indeed desirable, to develop a theory of the NPG that doescapture these realities and complexities. This theory is not integral to PA or to the NPM but is rather an alternative discourse in its own right. It ispredicated upon the existence of a plural state and a pluralist state, and itseeks to understand the development and implementation of public policyin this context.5

INTRODUCTION

As outlined above, therefore, PA is situated firmly within the political studies discipline. Influential theorists include Woodrow Wilson (1887) andWilliam Robson (1928). It has at its core a concern with the unitary state,where policy making and implementation are vertically integrated as a closedsystem within government. It focuses upon the policy making and imple-mentation cycle, with an assumption that effective PA is comprised of thesuccessful implementation by public managers of policies decided “up stream”in this system by democratically elected (and, it is implicitly assumed, account-able) politicians. Because of its vertically integrated nature, hierarchy is thekey resource-allocation mechanism for PA, with a focus upon vertical line-management to ensure accountability for the use of public money (Day andKlein 1987; Simey 1988). The value base is one based in an explicit assump-tion of the hegemony of the public sector for the implementation of publicpolicy and the delivery of public services.

Some writers, of course, have long recognized the fallibility of the PAparadigm without entirely dismissing it as a framework for the design and delivery of public services. The theory of “street level bureaucrats” (Lipsky 1979), for example, seeks to explain the breakdown of the “policy-maker–administrator” divide in conditions of resource shortage, but with-out dismissing in its entirety the framework of PA for the provision of public services (see also Schofield 2001 for a good overview of this range of arguments).

By comparison, the NPM6 is a child of neo-classical economics and par-ticularly of rational/public choice theory. Influential writers include Tiebout(1956) and Niskanen (1971). It is concerned with a disaggregated state, wherepolicy making and implementation are at least partially articulated and dis-engaged, and where implementation is through a collection of independentservice units, ideally in competition with each other. The key role of the state here is regulation, often within a principal–agent context (Vickers andYarrow 1988). Its focus is almost wholly upon intraorganizational processesand management.7 Drawing upon open rational systems theory, it models theproduction of public services as an intraorganizational process that turns inputs into outputs (services) within a mediating environment, and with anemphasis upon the economy and efficiency of these processes in producingpublic services. As already noted, it assumes competitive relationshipsbetween the independent service units inside any public policy domain, taking place within a horizontally organized marketplace – and where thekey resource-allocation mechanism is a variable combination of competition,the price mechanism and contractual relationships, depending upon whichparticular variant of the NPM one chooses to expound. Its value base is formedaround “the logic of accounting” and is contained within its belief that thismarketplace, and its workings, provides the most appropriate place for theproduction of public services. An extreme form of this argument is made byPirie (1988).

INTRODUCTION

In contrast to both of the above, the NPG, if it is to be situated as aparadigm of public services delivery, is rooted firmly within institutional andnetwork theory, and draws much from the influential work of Ouchi (1979),Powell (1990), Powell and DiMaggio (1991), and Nohria and Eccles (1992).It posits both a plural state, where multiple interdependent actors contributeto the delivery of public services, and a pluralist state, where multiple pro-cesses inform the policy-making system. Drawing upon open natural systemstheory, it is concerned with the institutional and external environmental pressures that enable and constrain public policy implementation and thedelivery of public services within such a plural and pluralist system. As aconsequence of these two forms of plurality, its focus is very much upon inter-organizational relationships and upon the governance of processes, stressingservice effectiveness and outcomes that rely upon the interaction of PSOswith their environment. The central resource-allocation mechanism is the inter-organizational network, with accountability being something to be negotiatedat the interorganizational and interpersonal level within these networks(Osborne 1997). Importantly, such networks are rarely alliances of equalsbut are rather riven with power inequalities that must be navigated success-fully for their effective working. Hence the value base in such networks isoften dispersed and contested.

The NPG is thus both a product of and a response to the increasingly complex, plural and fragmented nature of public policy implementation andservice delivery in the twenty-first century. Its key elements in relation to PAand the NPM are summarized in Table 1.1.

Now, significant work has already taken place that might legitimately be said to fall within the boundaries of the emergent regime of the NPG.This includes work upon the nature and governance of the policy process(Klijn and Koppenjan 2000, 2004), the issue of “managing outward” for PSOs and managers (Moore 1995; O’Toole et al. 2005), the development of keymanagement skills in an interorganizational context (Getha-Taylor 2008),expanding the nature and impact of accounting within PSOs to embrace amore holistic approach to their environment (Ball and Seal 2005; Marcuccioand Steccolini 2005), and the governance of interorganizational relationshipsthemselves (Hudson 2004; Huxham and Vangen 2005). Invariably, though,the focus has been at the organizational rather than the service system level.

However, it has become increasingly apparent that the public policy imple-mentation and public services delivery research agenda, certainly within theUK, is one where its parameters and questions have been set within the pre-vious regimes – and particularly within that of the NPM. This research agendais asking the old questions about public policy implementation and publicservices delivery. These questions are epitomized within the ESRC PublicServices Programme in the UK. This program has been important for theresearch community and has produced some outstanding research findings.Nonetheless, as a research agenda, it is characterized precisely by these “old

INTRODUCTION

questions” of intraorganizational efficiency and effectiveness. These old ques-tions can be summarized as:

• How do we manage public policy implementation to ensure that the polit-ical will is carried out in practice? (the policy implementation question)

• How do we ensure organizational and individual service performance?(the audit and targets question)

• How do we ensure that individual PSOs can work in partnership mosteffectively? (the partnerships question)

• How do we hold public managers accountable? (the scrutiny question)• How do we “incentivize” staff for optimal productivity? (the rewards

question)• How do we ensure organizational sustainability? (the change and inno-

vation question)

The argument here is that, if we are going to develop the NPG as a conceptualization of public policy implementation and public services management, it is necessary to move toward an integrated body of know-ledge about the NPG. This requires our research community to start askinga series of “new questions” about the fundamentals of the NPG. These questions are focused upon the underlying principles of public servicesdelivery in the plural and pluralist state and upon the public service system,rather than upon individual PSOs. These new questions are, it is argued here, sevenfold:

• What should be our basic unit of analysis in exploring public policy imple-mentation and public services delivery – and what are the implicationsof this for theory and practice? (the fundamentals question)

• What organizational architecture is best-suited to delivering public services in the plural state? (the architectural question)

• How do we ensure sustainable public service systems – and what doessustainability mean? (the sustainability question)

• What values underpin public policy implementation and services deliveryin such systems? (the values question)

• What key skills are required for relational performance? (the relationalskills question)

• What is the nature of accountability in fragmented plural and pluralistsystems? (the accountability question)

• How do you evaluate sustainability, accountability and relational per-formance within open natural public service delivery systems? (the evaluation question)

It should be emphasized that these new questions are not a simple replace-ment for the old ones. The imperative for effective intraorganizational and

INTRODUCTION

service management remains – and so the “old questions” still remain pertinent.However, such effectiveness by itself will not engender the delivery of suc-cessful public services in the contemporary plural and pluralist state. In orderto contribute to such delivery, the NPG needs to encompass both an activeresearch agenda that will explore the efficacy and limitations of the regimeand a developing body of capable and beneficial managerial practice withinthis plural and pluralist context. These questions will be returned to in theconclusions of this present volume, with a discussion of their implicationsboth for a research agenda and for policy practice.

Structure of this book

This volume is in five parts. The first part will provide a series of theoret-ical perspectives upon public governance and question the applicability of the concept of the “NPG”. Parts 2–5 will then explore the nature of public governance in a series of key, and indeed overlapping, areas: inter-organizational collaboration, contractual relationships, interorganizationalnetworks for services delivery, and public policy networks. The conclusionsto the volume will then consider the implications of public governance forpublic services delivery research over the next decade and beyond.

1 This chapter builds upon and expands the arguments made previously by this author in two prior pieces: S. Osborne (2006) “The New Public Governance?”, Public Management Review, 8 (30): 377–88, and S. Osborne (2009) “Delivering PublicServices: Are We Asking the Right Questions?”, Public Money and Management,29 (1): 5–7.

2 As one mainland European colleague remarked somewhat humorously to this authorrecently, the tradition in the UK is to see every change of national government asthe start of a new paradigm of public management and administration.

3 Ever since Hood’s influential essay on the New Public Management, there has been a tendency to herald every shift in public services provision as the “NewSomething-or-other”. Whilst there are clear limitations to this approach, it is nonetheless the one adopted here – primarily to differentiate it from the other diverse approaches to “governance” and “public governance” discussedbelow.

4 Brint Milward has made the point in a personal communication to this author thatthere is an inevitable national variance in the nature of public administration, andof the other regimes. The “politics–administration” split, for example, is perhapsa particularly European, and especially British, emphasis within PA. Moreover,all the elements are liable to buckle under extreme stress – the incrementalism ofPA was clearly put on hold during the two world wars of the twentieth century,for example.

5 These concepts of the plural state and pluralist state are reviewed further below.6 NPM, as discussed here, is very much the market-driven variant that emphasized

the efficacy of interorganizational competition in the delivery of public services.

INTRODUCTION

It is the model prevalent across the UK, the US, and Australia and New Zealandin particular. An alternative version, which is common across mainland Europe, does not place such an emphasis upon external competitive environment. Rather,it emphasizes contractual mechanisms within rather than without government(Schrijvers 1993).

7 Though Ostrom and Ostrom (1971) do offer a more explicitly interorganizationalapproach to public choice theory as a basis for the NPM.

References

Alford, J. and Hughes, O. (2008) “Public Value Pragmatism as the Next Phase ofPublic Management”, American Review of Public Administration, 36 (2): 130–48.

Ball, A. and Seal, W. (2005) “Social Justice in a Cold Climate: Could Social Account-ing Make a Difference?”, Accounting Forum, 29: 455–73.

Borins, S. (2002) “New Public Management, North American Style”, in K.McLaughlin, S. Osborne and E. Ferlie (eds) The New Public Management: CurrentTrends and Future Prospects, London: Routledge.

Börzel, T. (1997) “What’s So Special about Policy Networks? An Exploration of the Concept and Its Usefulness in Studying European Governance”, EuropeanIntegration online Papers (EIoP), 1 (16), http://eiop.or.at/eiop/texte/1997-016a.htm.

Chandler, J. (1991) “Public Administration: A Discipline in Decline”, TeachingPublic Administration, 9: 39–45.

Cornforth, C. (ed.) (2003) The Governance of Public and Non-profit Organisations.What Do Boards Do?, London: Routledge.

Dawson, S. and Dargie, C. (1999) “New Public Management: An Assessment andEvaluation with Special Reference to Health”, Public Management Review, 1 (4):459–82.

Day, P. and Klein, R. (1987) Accountabilities, London: Tavistock.Denters, D. and Rose, L. (2005) Comparing Local Governance, Basingstoke: Palgrave

Macmillan.Dunleavy, P. (1985) “Bureaucrats, Budgets and the Growth of the State”, British Journal

of Political Science, 15: 299–328.Entwistle, T. and Martin, S. (2005) “From Competition to Collaboration in Public

Services Delivery: A New Agenda for Research”, Public Administration, 83 (1):233–42.

Farnham, D. and Horton, S. (eds) (1996) Managing the New Public Services, Basing-stoke: Macmillan.

Ferlie, E., Ashburner, L., Fitzgerald, L. and Pettigrew, A. (1996) The New PublicManagement in Action, Oxford: Oxford University Press.

Flynn, N. (2002) Public Sector Management, London: Prentice Hall.Frederickson, G. (1999) “The Repositioning of American Public Administration”,

Political Science and Politics, 32: 701–11.Frederickson, G. and Smith, K. (2003) The Public Administration Primer, Boulder,

Colo.: Westview Press.Getha-Taylor, H. (2008) “Identifying Collaborative Competencies”, Review of

Public Personnel Administration, 28 (2): 103–19.Gow, J. and Dufour, C. (2000) “Is the New Public Management a Paradigm? Does

It Matter?”, International Review of Administrative Sciences, 66 (4): 573–97.

INTRODUCTION

Hanf, K. and Scharpf, F. (eds) (1978) Interorganizational Policy Making, London:Sage.

Hill, M. and Hupe, P. (2003) “The Multi-layer Problem in Implementation Research”,Public Management Review, 5 (4): 471–90.

Hood, C. (1991) “A Public Management for All Seasons?”, Public Administration,69: 3–19.

Hood, C. (1995) “The New Public Management in the 1990s: Variations on aTheme”, Accounting, Organizations and Society, 20 (2–3): 93–109.

Hood, C. and Jackson, M. (1992) “The New Public Management: A Recipe forDisaster”, in D. J. Parker and J. W. Handmer (eds) Hazard Management andEmergency Planning: Perspectives on Britain, London: James and James.

Hudson, B. (2004) “Analysing Network Partnerships: Benson Re-visited”, PublicManagement Review, 6 (1): 75–94.

Hughes, O. (2002) Public Management and Administration, Basingstoke: Palgrave.Huxham, C. and Vangen, S. (2005) Managing to Collaborate, London: Routledge.Kettl, D. (1993) Sharing Power: Public Governance and Private Markets, Washington,

DC: The Brookings Institution.Kettl, D. (2000) The Global Public Management Revolution, Washington, DC: The

Brookings Institution.Kickert, W. (1993) “Complexity Governance and Dynamics: Conceptual Explorations

of Public Network Management”, in J. Kooiman (ed.) Modern Governance,London: Sage.

Kickert, W. (1997) “Public Governance in the Netherlands: An Alternative toAnglo-American ‘managerialism’ ”, Public Administration, 75 (4): 731–52.

Klijn, E.-H. (2002) “Governing Networks in the Hollow State: Contracting-out, ProcessManagement or a Combination of the Two”, Public Management Review, 4 (2):149–66.

Klijn, E.-H. and Koppenjan, J. (2000) “Public Management and Policy Networks:Foundations of a Network Approach to Governance”, Public Management Review,2 (2): 135–58.

Klijn, E.-H. and Koppenjan, J. (2004) Managing Uncertainties in Networks, London:Routledge.

Kooiman, J. (1999) “Social–political Governance: Overview, Reflections and Design”,Public Management Review, 1 (1): 67–92.

Leftwich, A. (1993) “Governance, Democracy and Development in the ThirdWorld”, Third World Quarterly, 14: 605–24.

Lipsky, M. (1979) Street Level Bureaucracy, New York: Russell Sage Foundation.Lynn, L., Heinrich, C. and Hill, C. (2001) Improving Governance: A New Logic for

Empirical Research, Washington, DC: Georgetown University Press.McLaughlin, K., Osborne, S. and Ferlie, E. (eds) (2002) The New Public Manage-

ment: Current Trends and Future Prospects, London: Routledge.Marcuccio, M. and Steccolini, I. (2005) “Social and Environmental Reporting in

Local Authorities: A New Italian Fashion?”, Public Management Review, 7 (2):155–76.

Marsh, D. and Rhodes, R. (1992) Policy Networks in British Government, Oxford:Clarendon Press.

Meier, K. (1997) “Bureaucracy and Democracy: The Case for More Bureaucracy andLess Democracy”, Public Administration Review, 57 (3): 193–9.

INTRODUCTION

Metcalfe, L. and Richards, S. (1991) Improving Public Management, London: Sage.Milward, B. and Provan, K. (2003) “Managing the Hollow State: Collaboration and

Contracting”, Public Management Review, 5 (1): 1–18.Mischra, R. (1984) The Welfare State in Crisis, Brighton: Wheatsheaf.Moore, M. (1995) Creating Public Value: Strategic Management in Government,

Cambridge, Mass.: Harvard University Press.Niskanen, W. (1971) Bureaucracy and Representative Government, Chicago, Ill.: Aldine-

Atherton.Nohria, N. and Eccles, R. (eds) (1992) Networks and Organizations: Structures, Form

and Action, Cambridge, Mass.: Harvard Business School Press.Osborne, S. (1997) “Managing the Coordination of Social Services in the Mixed

Economy of Welfare: Competition, Cooperation or Common Cause?”, British Journal of Management, 8: 317–28.

Osborne, S. and Brown, K. (2006) Managing Change and Innovation in PublicService Organizations, London: Routledge.

Osborne, S. and Kaposvari, A. (1997) “Towards a civil society? Exploring its meanings in the context of post-communist Hungary” in Journal European SocialPolicy, 7(3): 209–222.

Ostrom, V. and Ostrom, E. (1971) “Public Choice: A Different Approach to the Study of Public Administration”, Public Administration Review, 31: 203–16.

O’Toole, L., Meier, K. and Nicholson-Crotty, S. (2005) “Managing Upward, Down-ward and Outward: Networks, Hierarchical Relationships and Performance”,Public Management Review, 7 (1): 45–68.

Ouchi, W. (1979) “Markets, Bureaucracies and Clans”, Administrative Science Quarterly, 25: 129–41.

Peters, G. (2008) The Two Futures of Governing: Decentering and RecenteringProcesses in Governing, HIS Political Science Series Paper 114, Pittsburgh, Pa.:University of Pittsburg Press.

Peters, G. and Pierre, J. (1998) “Governance without Government? RethinkingPublic Administration”, Journal of Public Administration – Research and Theory,8: 227–43.

Pirie, M. (1988) Privatization: Theory, Practice and Choice, London: Wildwood House.Pollitt, C. and Bouckaert, G. (2004) Public Management Reform: A Comparative

Analysis, Oxford: Oxford University Press.Powell, W. (1990) “Neither Market nor Hierarchy: Network Forms of Organ-

ization”, Research in Organizational Behaviour, 12: 295–336.Powell, W. and DiMaggio, P. (1991) The New Institutionalism in Organizational

Analysis, Chicago, Ill.: University of Chicago Press.Rhodes, R. (1997) Understanding Governance, Buckingham: Open University Press.Robson, W. (1928) Justice and Administrative Law, London: Macmillan.Salamon, L. (2002) The Tools of Government: A Guide to the New Governance, New

York: Oxford University Press.Schofield, J. (2001) “Time for a Revival? Public Policy Implementation: A Review

of the Literature and an Agenda for Future Research”, International Journal ofManagement Reviews, 3 (3): 245–63.

Schrijvers, A. (1993) “The Management of a Larger Town: Outcome RelatedPerformance Indicators and Organizational Control in the Public Sector”, PublicAdministration, 71: 595–603.

INTRODUCTION

Simey, M. (1988) Democracy Rediscovered: A Study in Police Accountability, London:Pluto Press.

Thatcher, M. (1995) The Downing Street Years, London: HarperCollins.Tiebout, C. (1956) “A Pure Theory of Local Expenditures”, Journal of Political

Economy, 64 (5): 416–24.Vickers, J. and Yarrow, G. (1988) Privatization: An Economic Analysis, Cambridge,

Mass.: MIT Press.Wilson, W. (1887) “The Study of Administration”, Political Science Quarterly, 2.

INTRODUCTION

Part I

THEORETICAL PERSPECTIVESON PUBLIC GOVERNANCE

GLOBAL PERSPECTIVES ON GOVERNANCE

Patricia Kennett

Introduction

The concepts of globalization and governance are firmly established withinpublic policy debates. However, the dynamics, nature and implications ofthe relationship between globalization and governance are fiercely contested.This chapter will begin by examining various definitions of the concepts and the linkages between them. It will then go on to consider the spaces and practices of governance under the condition of globalization. It will focus particularly on the emergence of the institutional structures of globalgovernance, the key actors, dynamics and practices of public governance.The remainder of the chapter will consider whether new forms and layers ofdecision-making and participation herald the arrival of a new multi-layeredpublic governance.

Governance and globalization

New forms of relationship and interaction between state and society, govern-ments and citizens, and state and not-state institutions have emerged in thecontext of what Jessop refers to as an arena of “unstructured complexity”(Jessop 2004) and Rhodes characterizes as “a differentiated polity” (Rhodes1997). These novel forms have been captured in the concept of governance(Swyngedouw 2005; Rhodes 1996; Daly 2003; Newman 2005; Kjaer 2004;Jessop 2004; Stoker 1998). Governance facilitates an understanding of theways in which power penetrates policy spaces, processes and practices, andthe formal and informal institutional arrangements which contribute to a“matrix of governance” (Lieberthal 1995) or “choreographies of governance”(Swyngedouw 2001), the ensemble of norms, patterns of behavior, networksand other institutions, and the power-play between them. As Koenig-Archibugi (2003) explains, within governance systems “problem solving is notthe preserve of a central authority able to impose solutions on subordinateagencies and individuals, but the result of the interaction of a plurality of

actors, who often have different interests, values, cognitive orientations, andpower resources” (p. 319).

Governance, as Daly (2003) asserts, is fundamentally about change. In muchof the literature, it is mainly concerned with the idea of a change from therealist perception of old government and the Westphalian system, wherebythe world is organized into territorially exclusive, sovereign nation-states, each with an internal monopoly of legitimate violence, to new governance.The former is characterized as having a governing administration which is regarded as the key dominant actor in the policy arena. Internally thereexists a clear hierarchy of authority, and governing is seen as a top-down,hierarchical process, with the nation-state taking centre stage. However, withthe change from government to governance, the governing administration is now only one player amongst many others in the policy arena. Thus, thepolicy arena has become visibly more crowded and contested, there are moreactors involved, the boundaries between the public and the private spheresare less precise, and the government’s command over the policy process issaid to have been transformed (Kjaer 2004; Newman 2005).

Globalization has become an essential context in which to locate the analysis of public policy. Whilst the concept has generated enormous debateregarding the substance, extent and nature of the phenomenon, it is inter-preted here as an economic, ideological, political and institutional projectfacilitated and enabled through a technological revolution, an ideological andpolicy shift, a strengthening of the role of international financial institutions,and a changing geopolitical landscape (Giddens 1999; Held and McGrew 2000, 2002). Increasing competition in global markets, pressure from inter-national institutions, domestic policy decisions to reduce national barriersto international economic transactions, combined with the impact of newtechnologies, “created the enabling conditions for the onset of globalisation”(Biersteker 1998: 24).

Globalization has given rise to a stronger and more influential institutionaltransnational and supranational element of a new public order which goesbeyond the traditional boundaries of the state. The concepts of global governance (Rosenau and Czempiel 1992) and, more recently, global publicpolicy (Stone 2008) have emerged to refer to the interaction of “multiple andfluid regimes and to . . . the role of actors, events, laws, and policies that fall outside the traditional boundaries of the state” (Macrae 2006: 527). ForRosenau (1995), “global governance is conceived to include systems of ruleat all levels of human activity – from the family to the international organ-ization – in which the pursuit of goals through the exercise of control hastransnational repercussions” (p. 13). Whilst Dingwerth and Pattberg (2006)argue that the term “global governance” has become a catch-all phrase for“almost any process or structure of politics beyond the state – regardless of scope, content or context” (p. 185), they state nevertheless that “the study of global governance acknowledges that a plethora of forms of social

THEORETICAL PERSPECTIVES

organization and political decision-making exist that are neither directed towardthe state nor emanate from it” (p. 191).

An important dynamic in understanding the relationship between glo-balization and governance is the recognition that, despite apparent differ-ences, governance is best-understood not only as a “general phenomenon”(Krahmann 2003: 223), occurring across different levels of analysis with governance arrangements at the national, regional and global levels displayingimportant similarities, but also as integrated and dialectical. Whilst globalpublic policy is distinct from the national process of policy-making, it is notdetached from it. Although national public institutions no longer serve asthe sole organizing center for policy in the context of networks and the riseof the new “relational rationality” as opposed to a traditional rule-based uni-versal rationality linked to the rise of the modern state and modern law (Ladeur2004), the state is by no means in retreat (Weiss 2005) or withering away.Rather, government power has been dispersed and reconstituted across newsites of actions augmented through new strategies and technologies (Jessop2004; Kennett 2008). As Sassen (2004) argues, the realm of the national andthe realm of the global are not two mutually exclusive domains. The stateis neither an autonomous actor constrained only by the structural anarchy ofthe international order, nor is it being dismissed as irrelevant in the contextof the emergence of a new cosmopolitanism. Drawing on Robertson’s (1995)earlier work, Holton (2008) refers to “methodological glocalism” as a wayof “retaining a sense of the significance of nation-states and institutions whilst also incorporating transnational processes in the global field” (p. 46).Rather than a single nested scalar hierarchy, Brenner (2001) refers to “a mosaic of unevenly superimposed and densely interlayered scalar geometries”(p. 606). For Holton (2008), the global and the national are co-present, and interact and intersect in a range of ways. According to Held (2000), weare witnessing a new regime of government and governance “which is dis-placing traditional conceptions of state power as an indivisible, territoriallyexclusive form of power”. Far from globalization leading to “the end of thestate”, it is stimulating a range of government and governance strategies, amore activist state (Held 2000: 422), and in some fundamental ways a newpublic governance.

The shaping of global governance

The idea of transnationalism and a transnational sphere of social action has emerged as a major way of understanding globalization and generallyrefers to a range of phenomena that are seen as transcending the boundariesor sovereignty of nation-states, such as the growth of transnational corpor-ations and the globalization of production, the emergence of a capital classand new social movements. The proliferation of NGOs, and their increas-ing visibility in world politics and policy, has generated powerful debate at

a time when traditional, nationally based channels of participation have beenin decline. For Lipschutz (1992), an increasing role for civil society in theglobal arena “represents an ongoing project of civil society to reconstruct,reimagine, or re-map world politics” (p. 291).

The shift to the globalization of production and finance represented a newsocial structure of production and power relations (Cox 2004), and an inter-national institutional structure that, according to Wilkin (2000), conformed“to the political-economic interests of, broadly speaking, the G7 core capit-alist states and their corporations” (p. 22). The emergence of this new socialstructure of production and power relations has been sustained through adominant-rationalist-knowledge structure (Cox 1987) through which “the activities of transnational corporations, transnational investment, globalrestructuring and the creation of global markets have been legitimized bydominant discourses of globalization, modernization, and social progress”(Steans 2004: 30). Rosenau (1990) refers to the increasing relevance of private“sovereignty-free actors”, multi-national corporations, transnational societies,and international governmental and nongovernmental organizations. TheUnited Nations Global Compact Initiative, for example, embraces and fur-ther promotes these developments through an ambitious example of globalcooperation among private companies, governments, multi-lateral organiza-tions and NGOs (Woods 2003), with over 4,700 corporate participants andstakeholders from over 130 countries. Actors from all sectors are broughttogether to “advocate and promulgate” ten core principles drawn from theUniversal Declaration of Human Rights, the ILO’s Fundamental Principleson Rights and Work, the Rio Principle on Environment and Development,and the United Nations Convention Against Corruption. It seeks to promotethese core values through the “development, implementation, and disclosureof environmental, social and governance policies and practices” (http://www.unglobalcompact.org).

However, as Sholte (2000) appropriately points out, it is the IFIs, the WorldTrade Organisation, and the OECD which have overshadowed agencies such as the ILO and UNESCO on questions of managing globalization. Thedominance of market-driven approaches to stimulate growth and competitive-ness has, according to Wilkinson (2005), “recast transnational corporationsinto mobilisers of capital, generators of technology and legitimate inter-national actors with a part to play in an emerging system of global governance”(p. 37). This is evident in the development of the World Economic Forum(WEF), “which brought together the top representatives of transnational corporations and global political elites” (Robinson and Harris 2000) and theTransatlantic Business Dialogue (TABD) (Farnsworth 2008).

The economic crisis of the 1970s had already increased dependence of many countries on international financial institutions such as the IMF andthe World Bank, and by the 1980s the expansion of capitalist markets andthe deregulation of financial markets, and the undermining of Keynesian

demand management and the “industrial settlement” established in many older Western nations marked the emergence of the political, ideological andinstitutional context of globalization. Consumer sovereignty was promotedand implemented through monetarism, the minimal state and new public management, and through notions of contract and choice. Pierre (2000) refersto these developments in Western democracies during the 1980s as representinga challenge to the state “from within” as the institutional arrangements, indus-trial settlement and citizenship regime of the postwar era were perceived asthe sources of poor economic performance, and barriers and impedimentsto effective competition within states and in the increasingly importantinternational markets. The neo-liberal agenda, which had had little influ-ence, increasingly took center stage, expanding from its initial stronghold ofChile, to Great Britain and the United States, becoming the focal point aroundwhich the logics of transnational diffusion were articulated.

An essential ingredient in establishing and legitimating the political, ideo-logical and institutional context was the rise of “economism”. For Kay (2008),the ascendancy of economic modes of thinking has been a key aspect of new governance structures and has impacted on the way that governance is constructed and practiced. He uses the term “economism” to refer to “governance structures where economic logic or economically inspired adviceis institutionally embedded, normalized and held as necessary in the deter-mination of policy choices” (p. 19). Carriers and channels of diffusion includeorganizational carriers such as the International Monetary Fund, the WorldBank and the World Trade Organisation, as well as private firms and multi-national corporations. These classical organizations have played a significantpart in the global spread of neo-liberalism, and more recently good govern-ance and active citizenship, as the “fracturing of economic consensus”(Thirkell-White 2007) has seen a shift from the discursive neo-liberalism andmarket fundamentalism of the 1980s to one of “inclusive neo-liberalism” (Porterand Craig 2004).

By the 1990s, governance and the context of diffusion was changing asneo-liberal fundamentalism, new public management and global institutionswere being called into question by civil society and national governments.Concerns began to emerge regarding the lack of accountability and trans-parency, the democratic deficit, the negative impact of IFI’s strategies andpolicies, particularly for people in less developed and emerging societies, andfragmentation in policy and provision resulting from the implementation of the New Public Management. In the national arena, particularly in GreatBritain, Germany and the USA, the changing political landscape saw the emergence of “Third Way” politics (Giddens 1994) and the “CommunitarianTurn”, the central themes of which included participative governance andpublic deliberation as well as “joined-up government” and interorganizationalcollaboration to address the diverse needs of the public. This was supportedby a discourse and policy agenda promoting active citizenship, decentralization,

local participation, self-help and partnership, and the development of newforms of governance.

The transnational business, political and intellectual elite, consisting of “globalizing bureaucrats, politicians, and professionals” as well as consumerelites from the media and commercial sectors, and chief executives of transnational corporations (Sklair 1995, 1998), has played a key role in transforming, strengthening and shaping the political and economic dimensionsof international institutions and networks. Thus, the global elite as well as hierarchical forms of global governance have been important conveyors ofideational norms and practices, discussed later in this chapter. In addition,as Robinson and Harris (2000) argue, “Studies on building a global economyand transnational management structures flowed out of think tanks, universitycentres, and policy planning institutes in core countries” (Robinson and Harris2000) and have been particularly successful in promoting a neo-liberal rationality (Cammack 2008) cohering around the themes of market liberal-ization, culminating in the well-known package of policy measures referredto as the “Washington Consensus” in the 1990s and, more recently, the “Post-Washington Consensus”, one of the cornerstones of which has been “goodgovernance”.

Global governance as a multi-actor perspective

Globalization and governance are interconnected dialectically in terms of scale(Brenner 2001), form, process and practice. Integrating the concepts analyticallyfacilitates an understanding of the changing relations of space and power(Macrae 2006: 52), and the ways in which power penetrates the spaces, pro-cesses and practices of governance. As Newman (2005) explains, “the imageof a hierarchical relationship between state and citizenry . . . is displaced by the idea of multiple parallel spaces in which power is encountered andnegotiated” (p. 4). Exploring the context, dynamics and discourse in andthrough which global governance has emerged and evolved highlights therelationship and interaction between different spatial scales and variousactors and institutions.

The concept of global governance engages with a multi-actor perspectiveand embraces a plurality of mechanisms that horizontally and vertically linkthe activities of various actors such as politicians, civil servants, internationalgovernmental organization committees and representatives, interest group and non-governmental organization representatives, and experts (Dingwerthand Pattberg 2006; Kennett 2008). For Kooiman (1993), the concepts of diversity, dynamics and complexity are central features of governance itself.The socio-political system is increasingly differentiated, characterized by multiple centers, and is one in which “Actors are continuously shaped by(and in) the interactions, in which they relate to each other” (Kooiman 2003: 2). Kooiman (2003) argues that it is the role of government to enable

interactions, encourage many and varied arrangements for coping with policy,and ensure equitable distribution of services between actors through self- andco-regulation and public–private partnerships, or what Kooiman calls “typesof societal interactions” and “modes and orders of governance”. However,for Kooiman, no single governing agency is able to realize legitimate andeffective governing:

No single actor, public or private, has all knowledge and infor-mation required to solve complex, dynamic and diversified problems;no actor has sufficient overview to make the application of neededinstruments effective; no single actor has sufficient action potentialto dominate unilaterally in a particular governing model.

(Kooiman 1993: 4)

Thus, there has been a shift from direct forms of governance to a processof governance exercised through a plurality of actors, sites, spatial scales, andprocesses, with an increasing reliance by governments on informal forms ofpower and influence rather than on formal authority.

Within this “differentiated polity”, a new form of coordination is requiredwhich, according to Rhodes (1997), emerges through networks. Within increas-ingly complex multi-level global governance, these networks, involving a morediverse range of actors, have become increasingly influential. Stone refers to the multiple and varied transnational policy spaces where global publicpolicies occur as “agora” (Stone 2008). The term is drawn from Greek history and refers to “a marketplace or a public square”, a place for social,political and economic interaction, where borders are ill-defined and fluid,which Stone (2008) equates with the relationship between the commercial andthe public domain in the modern global era. According to Stone, utilizingthe idea of the agora is useful

to identify a growing global public space of fluid, dynamic, and intermeshed relations of politics, markets, culture, and society. Thispublic space is shaped by the interactions of its actors – that is multiple publics and plural institutions.

(Stone 2008: 21)

According to Ruggie (2004), “the very system of states is becoming embedded in a broader, albeit still thin and partial, institutionalized arenaconcerned with the production of global public goods” (p. 500), or what Yeates(2008) refers to as “embedded transnationalism”. National public domainsof countries have become more interlocked, and essentially nationally provided public goods have become “globalized”, such as financial codes and standards, human rights, labor standards and health (SARS and HIV)(Kaul et al. 2003; Kaul 2005). The increasingly transnational character of a

range of policy issues has, according to Stone (2004), provided the rationalesfor “research collaboration, sharing information and co-operation on other activities that creates a dynamic for the international diffusion of ideas and policy transfer” (p. 38). Processes of interaction among the various actorsin international politics are now more frequent and intense, giving rise towhat Ladeur (2004) refers to as “flexible institutions” (p. 5) taking place beyondthe state. These “flexible institutions” have become increasingly influentialin the global arena, in contrast to the Westphalian international systemwhereby, as Ruggie (2004) points out, the public domain was constituted bystates that were “the decision-makers and executors of their joint decisionsand actions, which were authoritative to the extent they were so recognizedby states” (p. 505). The hierarchical international institutional architectureenshrined within the Bretton Woods system established after World War IIcomprised territorially distinct units engaging in “external” transactions.

As Holton (2008) argues, networks have become a characteristic modethrough which the transnational is organized. Thus, a crucial element of globalgovernance is the shift from markets and hierarchies toward networks andpartnerships as modes of coordination (Marsh and Smith 2000), involvednot only in influencing policy decisions but also in the business of governmentitself. Both the scope and the intensity of global networks have expandedrapidly since the mid-1980s. Holton (2008) describes networks as “forms ofmulti-social organisation that are distinct from . . . Markets and hierarchies”(p. 4), involving more enduring forms of social commitment and trust thanmarkets, but more flexible and less centralized than hierarchies. Global hierarchies are seen as undemocratic and remote, whilst markets are unableto respond adequately to the new environment.

Transnational governance is characterized by a move toward non-binding“soft” rules such as standards and guidelines, benchmarking and monitoring,in contrast to the Westphalian world where rule-making was expressed in“hard laws”. Djelic (2006) argues that compliance is increasingly reliant onsocietalization, acculturation, and normative pressures, access to membershipand resources as well as with the threat of sanction and through mechanismsof conditionality. In this context, social interaction through “various formsof direct interface and exchange that often takes place within and across social networks” (Djelic 2006: 70) has become an influential carrier of diffusion logics, as structuration and socialization processes define the rulesof the game and “reflect a particular ideology and associated practices” through which normative and symbolic systems, rules and ideological framesare established shaping behavior and interaction. As Scott (2003) argues, carriers are not neutral vehicles “but mechanisms that significantly influencethe nature of the elements they transmit and the reception they receive” (p. 879). Nor are they bounded and self-contained entities, but overlappingand multi-scalar. As MacRae (2006) argues, scales are not fixed but are bothfluid and interconnected in terms of processes and actors. Agents do not act

only within their respective policy spaces, and are therefore not confined toa singular role and identity and a singular spatial scale. The European Union,for example, provides opportunities for domestic actors in the policy processand vice versa. Transfer does not simply occur in a unilinear hierarchicalprocess from supranational to national to subnational but is ongoing andmultidirectional, and spatial scales are perceived as “overlapping areas ofpolicy, norms, values, power relations, and social interaction, where actorsare not confined to a single scale” (p. 528).

Within the transnational policy community, Stone (2008) identifies threetypes of “carriers of global policy processes involved in the diffusion of ideas,standards, and policy practice” (p. 30). The “internationalised public sectorofficial” operates in transnational executive networks. These are individualswho derive their authority from their official positions within their nation-state. The “international civil servant” is employed by an internationalorganization and, in theory, is impartial and non-partisan. However, the roleof these actors in terms of innovation and implementation in global policyis difficult to discern, but where national interests continue to be pursued theseindividuals have “considerable capacity to shape (or delay) policies becauseof their expertise, routines, and positions of power” (Stone 2008: 30). The“transnational policy professionals” represent the growing community of con-sultants, business leaders, scientific experts, NGO executives whose networkor association is often receiving public funding. These groups of actors interact within and across a range of policy networks, often playing differ-ent roles at different spatial scales, contributing to what Ladeur refers to asthe “broadening of the perspective of the participants through the commonalityof interchanging roles” (Ladeur 2004: 5). Thus, different types of networksmay overlap in both functions and activity, intensifying the diffusion of policyand discourse.

The limits to global governance

Global governance implies that, through various structures and processes,actors can coordinate interests and needs through multi-centered structuresand processes that are flexible and appropriate for achieving policy objec-tives and responding to policy issues on a global scale. However, there is noone mode of decision-making within the international arena, and indeed thetrend is toward an increasing fragmentation of governance architecture. Stone(2008) points out that this can lead to a “policy vacuum” in relation to the“ownership” of public problems and the delegation of responsibility.

Global environmental governance, for example, has grown “tremendouslyand uncontrollably”, with some 500 international agreements that nowinfluence the governance of environmental problems. The proliferation of multilateral environmental agreements, the development of a substantial body of international law, and the creation and diffusion of norms have few

equivalents in other areas of global governance (Green and Thouez 2005).However, some critics argue that the “crazy quilt” (Charnovitz 2005) of globalenvironmental governance has become “unmanageable in size and intractablein its complexity” (Green and Thouez 2005: 4).

Also evident is the creation of a growing body of international law concerned with the protection and enforcement of human rights. There are at least twenty-nine international conventions and declarations relatedto discrimination, including United Nations conventions on basic universalhuman rights and rights of excluded groups such as women, children andracial /ethnic minorities, as well as International Labour Organisation (ILO)conventions on the rights of workers, and declarations on race and racial prejudices (Marquez et al. 2009). As Spiro (2002: 3) argues:

International law in general is assuming greater consequentiality asa determinant of state behaviour. States are increasingly willing topass human rights agendas against other states, even in the absenceof geopolitical gains; and non-state actors, and interested publics,now garner power independent of states with which to advance compliance with international law.

However, it is still the case that the acts of states feature centrally in howthe human rights issues impact and are played out. The country of whichyou are a citizen must be a signatory of the particular treaty. Most humanrights conventions within the UN system do not create legally enforceablebinding obligations which individuals can impose on states. Mechanisms forprotection and enforcement take the shape of progress reports by countrygovernments, international conventions and meetings. Indeed, although a statemight be a signatory to the various human rights conventions, ratificationis permitted subject to reservations – for example on the ground that nationallaw, tradition, religion or culture are not congruent with convention prin-ciples. The Convention on the Elimination of All Forms of Discriminationagainst Women (CEDAW) adopted in 1979 by the UN General Assembly hasbeen particularly controversial, as the extended list of country declarations,reservations and objections to CEDAW indicates. According to the UNDepartment of Economic and Social affairs, “Some reservations are drawnso widely that their effect cannot be limited to specific provision in the Convention” (www.un.org/womenwatch/daw/cedaw/reservations.htm). Whilstnational governments have been increasingly willing to sign up to inter-national conventions and treaties, albeit with numerous reservations, thereis still a reticence not only to recognize the precise obligations that they have assumed in terms of social rights, but also to extend and strengthen theinternational human rights framework (Novitz 2008).

The global arena is a highly contested terrain, and fraught with asymmet-ries of power and tensions around particular policy issues. Brenner (2001)

refers to a “kaleidoscope effect” to refer to the qualitative differences in theorganization of scalar patterns and the nature of interaction at different spatial scales across policy areas. There are also problems relating to the implementation of public policy in a global context when both formal andinformal institutions of global governance often “lack both the authority and the means to enforce policy compliance” (Stone 2008: 27).

More recent concerns have emerged around the legitimacy, representative-ness and effectiveness of the IFIs, and the inadequacies of global financialgovernance and regulation in the context of the meltdown of global finan-cial markets and economic recession. Although these concerns were initiallyhighlighted following the Asian Financial Crisis, only limited reform wasenacted and, indeed, the role of international and supranational organiza-tions was enhanced through the provision of financial and technical assistance.During the current economic recession, the IMF has had an opportunity toreassert its role in the international economy in terms of immediate crisismanagement and with the long-term reform of the international financial system. At the time of writing, the IMF has agreed a $2.1 billion two-yearloan with Iceland, a $16.5 billion agreement with Ukraine and a $15.7 billionloan to Hungary. Other countries are currently negotiating potential loansfrom the IMF. The International Finance Corporation (IFC), the private-sector lending arm of the World Bank, has also announced that it will launch a $3 million fund to capitalize small banks in poor countries that arebattered by the financial crisis.

In terms of internal reform, the IMF allocates a country quota to eachmember based on the relative size of its economy. This determines the country’s financial contribution to the IMF, its voting power, and ability toaccess IMF financing. A two-year reform programme of the system of quotashares was introduced in 2006 to address their unfair, “misaligned” (IMF2008) distribution and to increase the voting share of emerging market anddeveloping economies: 135 countries will see increases in their voting power,with an aggregate shift of 5.4 percentage points (Figures 2.1 and 2.2).

Another central element of the reform package has been the attempt toincrease “the voice of low-income countries” through an increase in basicvotes. Basic votes are provided to all members on an equal basis. The agree-ment reached endorsed a tripling of basic votes, the first such increase sincethe IMF was established in 1945. Additionally, the Articles of Agreementwill be amended so that the share of basic votes in total voting power does not decline in the event of future quota increases. However, in spite ofreforms and financial assistance, the question of whether or not these keyinstitutions are “fit for purpose” was raised at the first international summiton the financial crisis held in November 2008. Issues to be addressed includedthe reform of the global financial architecture, strengthening coordination,consistency and quality among actors and regulatory standards, and increas-ing regulation and transparency of financial instruments and institutions. One

necessity for reform and improved international coordination. Although theEuropean Union has been slow to react, governments in Europe and beyondhave shown an increasing willingness to intervene in their economies. It isnational governments who have been the key players in the “socializa-tion of losses” (UNCTAD 2008) as “the threat of a meltdown has broughtgovernments back onto centre stage” (p. 1). The US administration instituteda $700 billion bailout in October 2008, and the US Senate accepted a further economic stimulus of around $900 billion in January 2009 (News Daily2009). In the UK a second bank-rescue package totaling at least £50 millionwas also announced in January 2009. Whether or not this increased enthu-siasm shown by national governments for intervention can be characterizedas the end of the neo-liberal consensus and a resurgence of Keynesianism,as some suggest, only time will tell as norms, procedures, practices and dis-courses of governance are developed and diffused across different spatial scalesand policy fora. What is clear, however, is that national governments haverecognized that, in a global world, national bailouts alone are inadequatestrategies not only for addressing the current crisis but also for promotingsustainable, equitable and more effective new public governance.

Conclusion

Internationalization is not a new phenomenon, but over the last thirty yearsthe nature and scope of transnational links and networks have broadenedand intensified. Governance in a global world is being shaped and reshapedin constellations of public and private actors that include states, internationaland regional organizations, professional associations, expert groups, civil society groups and business corporations. New political structures and policyspaces beyond the state have multiplied as new issues arise and networks ofactors mobilize to be involved. Innovative layers of governance, strategiesof operation have emerged, along with new technologies of governance.

Regulation has increasingly shifted from the national level to policy levelsbeyond the state, but this is not an indication of the demise of the state. Theshaping of global governance has often occurred as a result of state encour-agement or, as Thompson (2003) puts it, “under the shadow” of either hierarchy of market. As Jessop (2004) argues, there is still substantial scopefor the reconstituted state “to mediate between the increasing numbers of significant scales of action” (p. 18). However, within the differentiated myriad of networks of formal and informal institutions the increasing involve-ment of non-state actors in norm- and rule-making processes, and compliancemonitoring, confirms the arrival of a global public governance.

Contemporary global public governance is a complex, fragmented, unstableand highly contested arena. However, rather than anarchic and unruly, it isframed through norms, practices and discourses which are shaped, stabilizeand change over time. As Isin and Wood (1999) point out, late modernity

cannot be associated only with fragmentation and disruption but also withthe emergence of new social movements and of “numerous social, political,cultural and social groups in search of new kinds of identification, politiciza-tion and solidarity” (p. 154). Governance in a global age is characterized by novel fora and channels for participation and resistance as well as by new forms of stratification and exclusion. There are different conduits andspatial scales through which nations and peoples are enmeshed in the globalarenas, rights and claims expressed. In the context of an established yet stillevolving globalization and governance architecture, new forms of politics,representation and layers of governance are emerging, both internationallyand locally, constituting multi-layered public governance.

References

Biersteker, T. J. (1998) “Globalization and the Modes of Operation of MajorInstitutional Actors”, Oxford Development Studies, 26 (1): 15–32.

Brenner, N. (2001) “The Limits to Scale? Methodological Reflections on ScalarStructuration”, Progress in Human Geography, 25 (4): 591–614.

Cammack, P. (2008) “Poverty Policy and the Politics of Competitiveness”, in P. Kennett(ed.) Governance, Globalization and Public Policy, Cheltenham: Edward Elgar.

Charnovitz, S. (2005) “A World Environment Organization”, in W. B. Chambers andJ. F. Green (eds) Reforming International Environmental Governance: From Institu-tional Limits to Innovative Reforms, Tokyo: UNU Press.

Cox, K. R. (2004) “Globalization, the Class Relation and Democracy”, GeoJournal,60: 31–44.

Daly, M. (2003) “Governance and Social Policy”, Journal of Social Policy, 32 (1):113–28.

Dingwerth, K. and Pattberg, P. (2006) “Global Governance as a Perspective on World Politics”, Global Governance, 12: 185–203.

Djelic, M.-L. (2006) “Marketization: From Intellectual Agenda to Global PolicyMaking”, in M.-L. Djelic and K. Sahlin-Andersson (eds) Transnational Govern-ance: Institutional Dynamics of Regulation, Cambridge: Cambridge University Press.

Farnsworth, K. (2008) “Governance, Business and Social Policy”, in P. Kennett (ed.)The Handbook of Comparative Social Policy, Cheltenham: Edward Elgar.

Giddens, A. (1994) Beyond Left and Right, Cambridge: Polity Press.Giddens, A. (1999) Runaway World: How Globalization Is Reshaping Our Lives, London:

Profile Books.Green, J. F. and Thouez, C. (2005) Global Governance for Migration and the

Environment: What Can We Learn from Each Other?, Global Migration Perspec-tives No. 46, Geneva: Global Commission on International Migration.

Held, D. (2002) Regulating Globalization? In Held, D. and McGrew, A. (eds) TheGlobal Transformations Reader, Cambridge: Polity Press.

Held, D. and McGrew, A. (2000) The Global Transformations Reader, Cambridge:Polity Press.

Held, D. and McGrew, A. (2002) Governing Globalization, Oxford: Polity Press.Holton, R. J. (2008) Global Networks, Basingstoke: Palgrave Macmillan.

International Monetary Fund (2008) “Governance: Country Representation”, http://www.imf.org /external /about/govrep.htm [accessed 9 December 2008].

IPES (2008) Outsiders? The Changing Patterns of Exclusion in Latin America andThe Caribbean. 2008 Report Inter-American Development Bank.

Isin, G. E. and Wood, K. P. (1999) Citizenship and Identity, London: Sage.Jessop, B. (2004) “Hollowing out the ‘Nation State’ and Multi-level Governance”,

in P. Kennett (ed.) A Handbook of Comparative Social Policy, Cheltenham: EdwardElgar.

Kaul, I., Conceicao, P., Le Goulven, K. and Mendoza, R. U. (2003) (eds) Provid-ing Global Public Goods; Managing Globalization, New York: Oxford UniversityPress.

Kaul, I. (2005) “Private Provision and Global Public Goods: Do the Two GoTogether?”, Global Social Policy.

Kay, A. (2008) “Economism and Public Policy”, in P. Kennett (ed.) Governance,Globalization and Public Policy, Cheltenham: Edward Elgar.

Kennett, P. (2008) “Governance, the State and Public Policy in a Global Age”, in P. Kennett (ed.) Governance, Globalization and Public Policy, Cheltenham:Edward Elgar.

Kjaer, A. M. (2004) Governance, Cambridge: Polity Press.Koenig-Archibugi, M. (2003) “Global Governance”, in J. Mitchie (ed.) The Handbook

of Globalisation, Cheltenham: Edward Elgar.Kooiman, J. (1993) “Social–political Governance”, in J. Kooiman (ed.) Modern Govern-

ance, London: Sage.Kooiman, J. (2003) Governing as Governance, London: Sage.Krahmann, E. (2003) “National, Regional, and Global Governance: One Phenomenon

or Many?”, Global Governance, 9: 323–46.Ladeur, K.-H. (ed.) (2004) Public Governance in the Age of Globalization, Aldershot:

Ashgate.Lieberthal, K. (1995) Governing China: From Revolution through Reform, New York:

W. W. Norton.Lipschutz, I. (1992) “Reconstructing World Politics: The Emergence of Global Civil

Society”, Millennium, 21 (3): 300–420.Macrae, H. (2006) “Rescaling Gender Relations: The Influence of European Direc-

tives on the German Gender Regime”, Social Politics: International Studies in Gender,State and Society, 13 (4): 522–50.

Marsh, D. and Smith, J. J. (2000) “Understanding Policy Networks: Towards aDialectical Approach”, Political Studies, 48: 4–21.

Marquez, G., Chong, A., Duryea, S., Mazza, J., and Noro, H. (2009) 2008 Report.Outsiders? The Changing Patterns of Exclusion in Latin America and the CaribbeanInter-American Development Bank, Washington, USA.

News Daily (2009) “Government Add to Rescue Plan”, www.newsdaily.com.tre467me-us-financial/ [accessed 28 January 2009].

Newman, J. (ed.) (2005) Remaking Governance: Peoples, Politics and the PublicSphere, Bristol: Policy Press.

Novitz, T. (2008) “International Law and Human Rights in the Context ofGlobalization”, in P. Kennett (ed.) Governance, Globalization and Public Policy,Cheltenham: Edward Elgar.

Pierre, J. (ed.) (2000) Debating Governance: Authority, Steering, and Democracy, Oxford:Oxford University Press.

Porter, D. and Craig, D. (2004) “The Third Way and the Third World: PovertyReduction and Social Inclusion in the Rise of ‘Inclusive’ Liberalism”, Review ofInternational Political Economy, 11 (2): 387–423.

Rhodes, R. A. W. (1996) “The New Governance: Governing without Government”,Political Studies, 44: 652–67.

Rhodes, R. A. W. (1997) Understanding Governance: Policy Networks, Governance,Reflexivity and Accountability, Buckingham: Open University Press.

Robertson, R. (1995) “Glocalization, Time-space and Homogeneity”, in M.Featherstone, S. Lash and R. Robertson (eds) Global Modernities, London: Sage.

Robinson, W. I. and Harris, J. (2000) “Towards a Global Ruling Glass? Global-ization and the Transnational Capitalist Class”, Science and Society, 64 (1):11–54.

Rosenau, J. (1990) Turbulence in World Politics: A Theory of Change and Continuity,Princeton, NJ: Princeton University Press.

Rosenau, J. N. (1995) “Governance in the 21st Century”, Global Governance, 1 (11):13–43.

Rosenau, J. N. and Czempiel, E.-O. (eds) (1992) Governance without Government: Orderand Change in World Politics, Cambridge: Cambridge University Press.

Ruggie, J. (2004) “Reconstituting the Global Public Domain: Issues, Actors andPractices”, European Journal of International Relations, 10 (4): 499–531.

Sassen. S. (2004) “De-nationalized State Agendas and Privatized Norm-making”, in Ladeur, K.-H. (ed.) Public Governance in the Age of Globalization, Aldershot:Ashgate.

Scholte, J. A. (2000) Globalization: A Critical Introduction, Basingstoke: PalgraveMacmillan.

Scott, W. R. (2003) “Institutional Carriers: Reviewing Modes of Transporting Ideas over Time and Space and Considering Their Consequences”, Industrial andCorporate Change, 12 (4): 879–94.

Sklair, L. (1995) Globalization and World Society, Cambridge: Polity Press.Sklair, L. (1998) “Globalization and the Corporations: The Case of the California

Fortune 500”, International Journal of Urban and Regional Research, 22 (2): 195–215.

Spiro, P. J. (2002) “Mandated Membership, Diluted Identity: Citizenship, Global-ization, and International Law. Available at SSRN: http://ssrn.com/abstract = 322360[accessed on 28 April 2008].

Steans, J. (2004) “Globalization, the State and Welfare: Gendering the Debate”, inP. Kennett (ed.) Handbook of Comparative Social Policy, Cheltenham: Edward Elgar.

Stoker, G. (1998) “Governance as Theory: Five Propositions”, International SocialScience Journal, 50 (155): 17–28.

Stone, D. (2004) “Think Tanks beyond Nation States”, in D. Stone and A. Denham(eds) Think Tank Tradition: Policy Research and the Politics of Ideas, Manchester:Manchester University Press.

Stone, S. (2008) “Global Public Policy, Transnational Policy Communities, and TheirNetworks”, The Policy Studies Journal, 36 (1): 19–38.

Swyngedouw, E. (2005) “Governance Innovation and the Citizen: The Janus Faceof Governance-beyond-the-state”, Urban Studies, 32 (11): 1991–2006.

Thirkell-White, B. (2007) “The International Financial Architecture and the Limitsto Neoliberal Hegemony”, New Political Economy, 12 (1): 19–41.

Thompson, G. (2003) Between Hierarchies and Markets: The Logic and Limits ofNetwork Forms of Organisation, Oxford: Oxford University Press.

UNCTAD (2008) “The ‘Crisis of a Century’ . . .”, Policy Brief, no. 3, October, Geneva.Weiss, L. (2005) “The State-augmenting Effects of Globalization”, New Political

Economy, 10 (3): 345–53.Wilkin, P. (2000) “Solidarity in a Global Age – Seattle and Beyond”, Journal of World

Systems Research, VI, 1 (200): 20–65.Wilkinson, R. (2005) “The Commission on Global Governance in a New World”,

in R. Wilkinson (ed.) The Global Governance Reader, London: Routledge.Woods, N. (2003) “Global Governance and the Role of Institutions”, in D. Held and

A. McGrew (eds) Governing Globalization: Power, Authority and Global Governance,London: Polity Press.

Yeates, N. (2008) Understanding Global Social Policy, Bristol: The Policy Press.

META-GOVERNANCE AND PUBLIC MANAGEMENT

B. Guy Peters

The reforms that have been implemented in the public sector over the pastseveral decades have had a very wide range of motivations, and have hadan equally wide range of consequences for the public sector for the citizensof the countries in which they are being implemented. In almost any countryone can identify, the public sector is now significantly different from thatwhich was to be found several decades ago, and indeed in some cases thepublic bureaucracy would be hardly recognizable to civil servants who hadpreviously worked in government. The idea of many political leaders has beenthat the bureaucracy was the problem, not the solution, and that fundamentalchanges were required.

What is clear in all this change is that the traditional public sector is no longer the model of first and last recourse for individuals responsible forgoverning. The traditional model of bureaucratic governance involved a rangeof assumptions about how to govern, although the traditions of governingwere not the same in all countries. Walsh and Stewart (1992), for example,described that mode of governance for the United Kingdom. This modelemphasized the autonomy of the public sector, the separation of public-sector employment from the private sector, and hierarchical authority withinthe administrative system. Other administrative systems, such as in Franceand Germany, emphasized that hierarchical control perhaps even more, butalso accepted closer relationships between the administrative system and thecivil society, and political parties.

This chapter will first describe, if briefly, the changes that have been imple-mented in governing over these past decades.1 This description will argue that,although there are different styles of reform, the fundamental consequenceof these reforms has been to move governing out of the center of the con-ventional, politically driven public sector and to empower a range of actorsincluding (but not limited to) senior public servants, lower-echelon publicemployees, and members of civil society. The rhetoric and the reality of governing were transformed in this process, and many conventional stylesof governing were delegitimated.

These reforms, however, have produced a number of significant prob-lems in governance that I shall detail. Finally, I shall argue that there is anemerging style of governing from the center that can be described as meta-governance (see Sorenson 2006). It recognizes the need for some delegationand devolution of governing but at the same time recognizes the need forgreater central direction. This paper will take the classic newspaper reporter’sapproach, and ask “Who, What, Where and Why” about meta-governance,and especially about meta-governance directed at devolved public organiza-tions and to networks. Meuleman (2008) and others have also documentedthe need to consider meta-governance of more conventional components ofthe governance process, but I shall focus more on the rather recent changesin the administrative process. Further, I shall invest more time and space in explaining the “why” of creating meta-governance, given that this aspect of the story being told here is crucial for understanding the choices beingmade by politicians and by public servants.

What is meta-governance?

I shall be using the term “meta-governance” to describe the process of steer-ing devolved governance processes. In other terms, it is the “governance ofgovernance”. The notion of meta-governance is that a number of organiza-tions and processes within the public sector have attained a substantial degreeof autonomy – a condition often described as governance – and that theremay be a need to impose some control over those components of governing.As argued below, the changes within the public sector during the past severaldecades have been a major source of the perceived need for meta-governance;but in some cases, e.g. the United States (Carpenter 2001), there have beensignificant requirements for meta-governance within the political system forsome time.

Whether the administrative process has been altered because of the accept-ance of managerialist reforms or because of more participatory styles of governing is largely irrelevant; the real or perceived requirement for meta-governance will be roughly the same. That requirement will be to providedirection to the administrative system, but to do so through mechanisms that maintain the virtues that have been produced by delegated and devolvedforms of governing, while providing central direction and control. Governinghas always involved some balancing of control and autonomy for public organizations, and for individual public servants, but that balance becomesmore apparent when decisions must be made about reasserting greater management controls over devolved systems.

As governance has become of much greater importance in the academicliterature on the public sector, as well as in the real world of governing, scholars have begun to develop the concept of meta-governance. For example,Louis Meuleman (2008) has provided the most extensive discussion of the

concept of meta-governance now available, focusing on its relationship tothe three conventional modes of governance: hierarchies, markets and net-works. Bob Jessop (2002; see also Bell and Park 2006) has also provided aseries of discussions of meta-governance, focusing attention on governancefailures and the need to develop strategies for coping with those failures. Also, Sorenson (2006) has emphasized the need to reassert political controlsover the devolved governance processes.

The available literature does an adequate job of developing the conceptof meta-governance, but has been less effective in examining the instrumentsthat may be available for providing the steering needed. To some extent, the instruments available to would-be meta-governors are much the same asthose available to any actor available in the public sector, that range of instru-ments described by Christopher Hood (1976) as NATO (nodality, authority,treasure and organization). These tools are almost entirely generic, so maybe adapted to cope with the meta-governance requirements, but they do notprovide any specific analytic advantage for understanding the issues. The logic of the tools of the “new governance”, as described by Salamon (2001),are more closely aligned with the needs for meta-governance, but still do not provide specific assistance for meta-governance.

The particular issue differentiating instruments for meta-governance from other instruments for governing is that these “tools” are directed atthe internal control of the public sector itself, rather than at influencing theeconomy and society. While much of the same logic is involved, there maybe some important differences as well. In particular, when attempting to imposecontrols over organizations within government itself there is an implicit, orexplicit, understanding that those organizations should accept the authorityof the controllers. This assumption may be incorrect, and many public organizations do attempt to evade authority, but the “shadow of hierarchy”is clearly darker within government than without.

Finally, meta-governance can be conceptualized as directed at controllingthe environment of action in the public sector, rather than controlling thataction directly. Theodore Lowi (1972), for example, argued that controllingthe environment of action could be used more generally to classify publicpolicies.2 Further, if the would-be governor is faced with difficult more-control situations, attempting to frame action indirectly may be the most effective manner of governing. For example, regulations tend to shape theenvironment of public action, as has been further elaborated through regu-lation theory (Scott 2004).

Reforms: toward the decentered state

The numerous and varied reforms in the developed democracies are difficultto summarize, but in this chapter I shall focus on two of the more import-ant strands of change (see also Peters 2001). As already noted, these two

versions of reform have tended to move the public sector away from the moreor less bureaucratic nature of governing. They both attempt to find ways ofinvolving a wider range of actors – from the market and from civil society– in the processes of governing. These transformations, unlike many otherattempts at reform in the past, involve a clear theory of governing and ofthe public sector, and are also more coherent than many previous attemptsat reform. That said, there are several alternative conceptions which havesome common elements but which also have rather different assumptionsabout the way in which the public sector functions.

The most familiar of these reforms have been captured in the phrase New Public Management (NPM).3 The basic idea of these reforms has beento transform the public sector so that it looks and acts a good deal morelike the private sector. Hood (1991) and others (Christensen and Laegreid2007) have pointed to a number of internal differences within NPM,although at the same time recognizing that there is some core to the patterns of change. The use of market principles within NPM has been conceptualized as a means of enhancing the efficiency and effectiveness ofthe public sector, by opening up the public sector to more competition andusing more business-like instruments for management. That competition was to be from outside government as well as among organizations andemployees within government.

As well as being an approach to management, the NPM is a theory ofgoverning more generally. One of the fundamental arguments, if generallymade implicitly, is that political domination of governing is likely to beinefficient and therefore governing will be improved if as many things as possible are removed from the hands of the political class. The tendency tomove activities out of the center of government into agencies, and to use a variety of alternative service-delivery mechanisms, has tended to reducethe number of levers that political leaders can pull to achieve the results they desire, and the results they promised in elections. Those structural andprocedural reforms, in turn, appear to have allowed managers to make moredecisions about the content of policy as well as about its implementation.

The principal alternative to the NPM has been a series of participatoryreforms designed to enable not only citizens but also lower-level public servants to have greater influence over policy and implementation. Thesereforms have been justified largely on democratic terms, in contrast to theefficiency and effectiveness associated with NPM. That said, a number ofscholars of public administration, as well of organizations more generally,have argued that involving the lower echelons of organizations and membersof civil society more directly in decisions provides more information aboutthe actual functioning of programs.

One important extension of the participatory reforms has been the develop-ment of “governance”4 as the means for delivering services, and often fordeciding on policy, in a number of developed democracies. The logic of the

governance reforms has been that the conventional hierarchical models ofbureaucracy, and the conventional representative models of democracy, havebecome outmoded. The public appear to be losing faith in political partiesand in elections (Dalton and Wattenberg 2000; Mair and van Biezen 2001),so that contemporary states are increasingly legitimated by their outputs ratherthan by the democratic processes that have in the past legitimated them. In addition, the bureaucratic processes through which governments have delivered public services are seen as clumsy, ineffective and unresponsive.

Those conventional models of governing have not recognized their ownfailures, nor the development of elements in the civil society that are cap-able of providing a substantial amount of self-steering in their own policyareas. As networks of social actors have come to take more responsibilityfor aspects of governing, the emergence of the concept, and the reality, ofgovernance has meant that the state can be relieved of many of its more difficulttasks. Some scholars (Rhodes 2000) have made the extreme argument thatsocieties can be governed almost entirely by self-organizing networks of social actors. Even in less extreme views, however, it is clear that social actors– interest groups, NGOs, and perhaps some individuals – are providing governance in a wide range of policy areas.

The governance argument, and again its reality, is not viable in all polit-ical systems. The heartland of this style of governing is Northern Europe,especially Scandinavia and the Low Countries, although there are notableexamples from the United States (Milward and Provan 2000). The dicho-tomy between state and society in the Anglo-Saxon world has made this styleof governing more problematic, and often the close relationship between the public sector and groups may be considered corruption. Further, in theless-developed world and in transitional regimes there may not be a civil society with sufficient strength to provide an alternative to formal governancethrough the public sector. Despite these caveats, there has been a sense thatthe basic logic of network governance has become ever more widely spread,in practice as well as in academic discourse.

The need for meta-governance

The reforms of the public sector associated with both NPM and “governance”have tended to produce a number of benefits but have also produced a num-ber of unintended and largely negative consequences. These consequencestend to affect not only the management of public programs but also the political process itself. One can identify a wide range of problems arising fromnetwork governance, but I shall discuss only four of the more significant ones here: decision-making, coordination and accountability. Also, while Iam discussing these problems as if they are self-evident, they may not be;and a good deal of the politics of contemporary network governing reflectsattempts to overcome these issues.

Decision-making

The rather clumsy bureaucratic institutions associated with conventional governing do have their problems but they do have the capacity to makedecisions. In representative democracy, majority rule and other constitutionalconventions ensure that decisions can be made.5 Likewise, implementationthrough public bureaucracies is done following legal constraints, so that againthere can be decisions. These words of praise do not mean that the decisionstaken are high-quality decisions, only that a decision can be made.

Making decisions in the networks associated with “governance” is not as easy as it is sometimes assumed to be. Unlike formal institutions, mostnetworks, or other types of structures involving social actors in governing,do not have clear decision-rules. The implicit norms are to bargain to con-sensus, but that bargaining may produce decisions by the lowest commondenominator (Scharpf 1988) unless there are other norms in place concern-ing checking one’s self-interest at the door.6 Without either ex ante rules or strong informal norms, governance is likely to produce no decisions, orrelatively poor quality decisions. While the advocates of network govern-ance may be excessively optimistic about the self-organizing nature of thosestructures, they do seek to provide an alternative to hierarchical control.

The above discussion of networks can be extended to include some market-oriented versions of service delivery associated with NPM. For example, con-tracts and public–private partnerships involve agreement among the partiesinvolved, and can therefore result in no deal being struck, or in a deal beingmade that does not move much beyond the lowest common denominator.Further, part of the logic of NPM has been to infuse the public sector wthprivate-sector values, and therefore the decisions being made may not reflectas strong a sense of the public interest as might be desirable.7 Again, to makethese arrangements function in the public interest may require infusing themwith values that go beyond the interests of the separate parties involved.

Participation

A major democratic issue involved in the development of governance throughnetworks and other more informal mechanisms is simply whether demo-cracy is actually advanced through mechanisms that are purportedly moredemocratic. Part of the logic of developing networks of social actors is thatconventional representative democracy is failing as a means of translatingpublic wishes into action. If nothing else, there is substantial evidence thatpublic involvement with political parties and in conventional modes of par-ticipation has been declining markedly (Mair and van Biezen 2001). Further,networks are designed to involve more actors in processes of governing and also to involve them in governing on a more continuous basis, so thatdemocracy is not just at election time.

Despite the problems inherent in representative democracy, it is not clear that networks will do that much better in representing social interestsin governing. The network models of democracy depend upon the involve-ment of the full range of interests and individuals in society, but the verypeople who may be excluded from effective participation in representativeinstitutions may also be excluded from participation in the networks. Beinginvolved in a network for governance involves, at a minimum, some level oforganization; and the socially excluded in societies are, almost by definition,those with minimal organizational skills. Even if somehow included in net-works, those same people would have, on average, fewer persuasive skillsand would, again on average, be less capable of influencing decisions.

Coordination

Governments have had difficulties in coordinating their activities as long asthere have been governments. Individual organizations and programs pursuetheir own goals, often at the expense of broader goals for government as awhole. The absence of effective coordination in the public sector generallyreduces the efficiency and effectiveness of public programs, and also enhancesthe familiar public perception of the incompetence of government. Further,a failure to coordinate effectively will produce lacunae, with some importantpublic problems not being addressed adequately.

Although poor coordination is a familiar problem, it has been exacerbatedby the reforms of the past several decades. For example, the use of the agencymodel for service delivery has created an even greater number of autono-mous organizations that then have to be coordinated. Likewise, networksand public organizations closely linked with networks may be more difficultto coordinate than organizations within the public sector. Finally, senior man-agers have been empowered to make more of their own decisions, so that they,too, may be more difficult to coordinate through the usual mechanisms.

The general strategy for decentering reforms has been to focus on goodmanagement and efficiency within a single organization or policy area. Thatstrategy has had numerous successes, but at the same time it weakens thecapacity of governments to impose general policy initiatives and priorities.For example, the creation of a large number of autonomous or quasi-autonomous agencies has created many more organizations that need coordinating.

Accountability

Finally, the principal issue that arises in the development of NPM and ofgovernance styles in the public sector is accountability for the decisions thatare taken in the name of the public. Both these alternatives to traditionalforms of governing tend to attenuate the linkages between public action and

political organizations, and to assume that those political linkages are perhapsless important than had been thought conventionally. There are further assump-tions about alternative forms of accountability (see Mulgan 2000, 2009) andcontrol, but these alternative mechanisms may not substitute in a demo-cratic system for the primacy of politics and ministerial responsibility for public action.

Accountability is often conceptualized as a mechanism for enforcing control over public organizations and programs, but it is also a means ofguiding the improvement of programs. The structural and procedural ele-ments contained in the reforms over the past several decades not only weakenthe capacity to control government; they also limit the error-correction elements of accountability. The logic is much the same, with the breakingor attenuation of the linkages between the political leaders and program delivery implying that any learning or control also will be broken.

Summary

The use of both NPM and the governance approaches to organizing and managing the public sector have produced some clear benefits for the public sector and for citizens; but, as with any set of reforms, there are alsocosts involved. These costs have been minimized by the advocates of thesereforms, and have also been less obvious because there have been some realbenefits created and also because these reforms were compatible with the general neo-liberal ideology of the era in which they were implemented. Inaddition, the reform process has been supported and advocated by a range ofinternational organizations that have legitimated and diffused these changesin the public sector.

It is to some extent obvious what problems have been created by thesereforms, but it is less clear what the most appropriate solutions to these prob-lems would be. The simplest response to these changes might be to attemptto return to the status quo ante. That is almost certainly not possible, althoughprobably more for the participatory and network reforms than for the NPM reforms. Having permitted greater involvement of both employees andsocial actors, it would be difficult to eliminate that participation. Therefore,political or administrative leaders who want to be able to steer effectivelymust find ways of using the gains achieved through prior reforms while stillovercoming the problems that have been created.

Meta-governance strategies: when should we meta-govern?

The final question to be considered when dealing with the meta-governanceof public programs is when, within the policy process, to emphasize meta-governing. The traditional approach to public management relied oncontrolling administrative processes and using ex ante rules to ensure that

decisions conform to law and to the directions coming from the political leadership. That hierarchical style of governing has been the target of a great deal of the reform efforts already described. The reforms described above have tended to reverse much of the hierarchical nature of governing,permitting individual public organizations as well as networks to makemore of their own decisions and to be held accountable for those decisionsafter the fact.

The implications of the changes in management are that any attempts to restore control over implementation will have to be ex post, attempting to detect problems after the fact and asserting control largely through errorcorrection. To some extent, accountability has always relied upon error correction and punishment, and even more contemporary instruments forcontrol such as performance management (see below) may depend heavilyupon detecting poor performance after the fact. That error correction is valuable but may also be inefficient in permitting the misuse of resourcesand the waste of time. Therefore, attempts at meta-governance that only correct errors should not be dismissed entirely, but they are perhaps less efficient at altering steering than are more direct instruments. Governmentsare far from perfect in governing, as an extensive literature documents, but having them overseeing other actors still provides for redundancy in controlling policy.

Instruments for meta-governance

The choice of instruments for meta-governance represents an answer to the question of “How” to meta-govern. Instrument choice is an important question for any attempt at governance, but is perhaps more difficult for meta-governance because of the numerous challenges already discussed. Thetask is to find instruments that can provide effective steering while at thesame time permitting the targets of the control to retain a good deal of theircapacity to make autonomous decisions concerning a range of activities.

Performance management

Performance management has been one of the most common instrumentsassociated with NPM. The basic logic of performance management is todevelop suitable measures of the outputs and outcomes of public action, andby using those measures to drive improved service delivery by the public sector. Making this method effective in turn requires the identification of thegoals of government and making the linkage between those goals and theindicators of public-sector activity. Although there are numerous problemsin the identification of appropriate measures, the basic idea can be help-ful in both managing organizations and in holding those organizationsaccountable.

Performance management can also be seen in the context of meta-governance.Setting the goals enables senior managers and/or political leaders to deter-mine what the organization should be doing, while allowing the organizationsto shape the actual means through which those goals would be reached. Thelatitude that the organizations may actually have in that implementation mayactually be constrained to some extent by the goals, but even then there issome ability for a public organization or a network of actors. If performanceis indeed what matters, then the organizations and networks should be able tomake many of their own decisions about the actual provision of the services.

Strategic management

Much of traditional management within the public sector has been detailed,and has tended to specify procedures and the details of programs. Further,much of that management has been conducted within individual programsand organizations, and, even if successful in improving performance, doesnot contribute to overall governance of a political system. Indeed, enhancedperformance within the single organizations may reduce the overall per-formance of a system by reducing the probability of coordination among thevarious components of the system.

Therefore, one important strategy for improving the governance of thatentire system of governance is to focus on policy coordination. The focuson coordination, in turn, should proceed from negative coordination throughto positive coordination (Scharpf 1994) and finally through to strategicmanagement. The more strategic approach would continue to emphasize coordination among the actors, but would do so around the principal goalsfor the political system and the society as a whole. This would be in contrastto beginning with the limited, disparate goals of the individual organizationsand programs, and attempting to build coherent policies from them.

Strategic management can be thought of as meta-governance because,although it posits the central values and goals for the policy process, it doesnot need to specify the means of achieving those goals. As already noted for performance management, establishing goals and then permitting choicesabout attaining the goals provides substantial controls over policy directions,and even about the style of implementation, but also preserves some aspectsof the autonomy of organizations and networks. If, indeed, this meta-governance functions within the shadow of hierarchy, then the center canalways pull the power back and impose more direct controls; but the initialefforts at meta-governing could be more indirect.

Budgets, personnel and the golden thread

In addition to the other reforms implemented within the public sectors, anumber of political systems have engaged in extensive internal deregulation

(see DiIulio 1994). This deregulation has weakened rules in areas such asprocurement, personnel management and the budget. Further, public bud-geting more generally has been reforming to provide managers with greaterlatitude in making decisions about the use of money. “Bulk budgeting” provides managers with a global budget, which they can use rather freely to make their programs perform as well as they can, within the bounds of law(see Scott 1996).

Contracting and the use of networks also provide a great deal of man-agerial freedom. The managers responsible for monitoring contracts and partnerships are also being given substantial latitude, and require that latitude to be able to be effective in more complex governance situations.Given that almost necessary latitude for managers, maintaining some threadof control is essential for asserting the primacy of political control over the activities of the public sector. The control need not be extensive but still is necessary, so therefore relying upon a relatively simple instrument or indicator may be crucial for meta-governing and much less disruptive thanmore direct instruments for control.

While controlling the budget is perhaps the most obvious means of meta-governing, controlling other basic inputs into the system may accomplish the same purpose. For example, controlling personnel allocations can alsoprovide a relatively simple source of control over individual organizations,albeit perhaps not as direct as it might have been when a greater share ofpublic-sector activity was provided directly, by career public servants. Controlsover secondary legislation also function as powerful means of controlling seemingly autonomous public organizations, although these are less effectivefor networks or other more devolved forms of public governance.

Soft law

Much of the process of governing is conceptualized as the use of law andformal authority. This style of governing has a number of positive featuresand has served most governance systems very well. However, just as increas-ingly true for governance (Salamon 2001) more generally, these formalinstruments appear less acceptable to people working within government. Muchof the logic of the NPM has been that public managers should have greaterindependence from these types of controls, while the logic of the more participatory style of governing has been that lower levels of government should also have greater autonomy. Both of these approaches therefore havetended to lessen central controls without replacing them with establishingcompensatory governance systems.

The idea of soft law has been applied primarily in international governance(Abbott and Snidal 2000) and within the European Union (Mörth 2004). Inthese settings there has been a marked absence of formal authority and legitimacy that is central to traditional institutional settings of governance.

The European Union has been able to use soft instruments to address policyareas that are formally outside its competence. By shifting from formal instru-ments associated with the “Community Method” to the softer instrumentsbased on negotiation, goals were reached that could not have been reachedthrough the more conventional means. The Open Method of Coordinationin the EU (Borras and Jacobssen 2004) and the European Social Dialogue(Smismans 2008) are two of the more evident examples of the use of softlaw in the multi-national setting.

The lessons derived from the experiences in international settings areapplicable to the national and subnational levels. While these governmentsmay be functioning within their area of competence, they may still confrontan increasing range of barriers to the use of formal instruments for control.Certainly networks and other forms of informality in the public sector pre-sent resistance to hierarchical controls, as do autonomous organizations withinthe public sector itself. These organizations also tend to have a stronger polit-ical base than do those in the international environment, and hence may bemore capable of resisting efforts to impose stronger forms of control thanare the organizations in the international setting.

Perhaps the most important difference between actions within the inter-national system and national policy-making is that there is at least some vestige of a shadow of hierarchy at the national level, regardless of the decisions that have been made that tend to lessen the level of control withinthose largely hierarchical systems. Thus, while soft law may be the only optionfor many dimensions of international governance, for domestic policy softlaw may just be a convenient and efficient instrument, or set of instruments,for coping with potential political resistance to imposition of central policydecisions.

Trust and values

The above instruments for meta-governance depend upon relative formal-istic and structural approaches, albeit ones that do not rely exclusively (if at all) on hierarchy. Using formal solutions is the natural reaction for mostpeople working in the public sector, but a softer approach to governing (ormeta-governing) may be to attempt to build values that can steer the rangeof organizations. There may be some common political values in societies(Painter and Peters 2009), as well as common administrative values (Hofstede2001), that can shape the implementation of public programs and facilitateachieving public purposes.8

The use of trust and values corresponds to the idea of shaping behaviorthrough shaping the environment of action described in the introduction.Governing through these instruments represents one of the cheapest and mosteffective approaches to meta-governance, and to governance more generally.If the would-be meta-governors can shape the values and the incentives to

which the individuals making decisions respond, then the outcomes desiredmay be reached with little investment of resources, and with a continuingeffect. The difficulty comes, of course, if the meta-governors change – e.g.there is a change in the governing coalition and there is a new set of valuesto guide decisions.

In the context of network governance and other more informal styles of governing, the use of values to steer requires to some extent that thosevalues represent strongly the importance of the public interest. Scholars suchas Moore (1995) have emphasized the importance of public values withinthe public sector itself, but those same values need to be extended to the otherstructures providing public services. Indeed, those values are more crucialfor these devolved structures that are inherently more impacted by the values of the private actors (market or non-market) that are containedwithin them. Further, if these basic values about the publicness of the public sector are institutionalized well, then even changing the operatingpremises of policy – again, as when there is a change in government – canbe accomplished more readily.

Summary

Meta-governance, and the perhaps more basic concern with governance, reflectsthe need to provide direction to economy and society, as well as the diffi-culty of providing that steering. Attempts to provide less direct mechanismsfor steering society, and granting greater autonomy to actors within the public sector, have generated numerous managerial benefits. These reforms,however, have also created managerial problems within the public sector. These reforms have also created many political problems by reducing thelevels of control that political leaders can exercise over public policies.

While reducing the levels of direct political involvement in the details ofpolicy may have been desirable from the perspective of increasing efficiency,it was less desirable from a democratic perspective, and many politicians (andtheir constituents) have sought greater control by those elected representatives.As already noted, meta-governance strategies provide a way of providing that direction while at the same time preserving some of the efficiency gainsfrom decentering reforms of government. The choice of meta-governance strategies, then, is an attempt to reassert some balance of power within thepolicy-making systems of the public sector, and to continue to involve non-state actors in the process while recognizing the primacy of politics.

One administrative reform tends to beget the next, as the difficulties created by one set of changes tends to create demands for additional changes.The question now is what will be the reaction to meta-governance reformsthat are being implemented. One option might be to make the state role ingovernance even stronger, especially when having to confront the probabil-ity of a long and severe economic downturn. The other option might be to

find new ways of involving the market and civil society in governing and toweaken further the role of the political aspects of governing.

These choices will not be easy, and may be country-specific. Much of theresearch on reform in the public sector has tended to assume rather commonpatterns of change, but we may now expect somewhat greater divergenceamong these systems. Coping with a major economic crisis, as well as withother major policy challenges such as climate change, may cause states torevert to more traditional patterns of governing. That would mean that moreliberal states would rely heavily on the market, and more étatiste states wouldreassert more direct state dominance. The observed pattern to date has been to some extent the reverse of that, with liberal Anglo-American systemsbecoming heavily involved with steering and some European states such as Germany being more laissez-faire. Still, coping with a crisis can be a spurfor reform and can help reveal future patterns of governing.

1 One suitable date for marking the beginning of the transformations would be theelection of Margaret Thatcher in 1979. See Savoie (1994).

2 In many ways, Lowi was writing about meta-governance without doing so expli-citly. See Nicholson (2002).

3 These reforms are hardly new, they emphasize the private more than the public,but they are indeed about management.

4 Governance is used to describe a variety of different aspects of the general processof governing. Here it is being used in a somewhat restrictive sense to mean the useof networks of social actors to do some parts of the work of governing that mightonce have been done by formal public organizations.

5 The major exception to that generalization would be presidential systems in whichthe conflicts between relatively equal branches of government can produce “grid-lock”. Some of the same gridlock may be produced in parliamentary systems thathave the possibility of being “divided” because of second chambers (Elgie 2001).

6 Network structures may therefore reflect the classic problem of the commons, inwhich the pursuit of individual interest may undermine the collective good. Theevolution of rules and norms as described in Ostrom’s work (1996) appears to represent the exception rather than the rule in these cases. In particular, the casesof developing rules autonomously tends to occur primarily in smaller settings ratherthan in situations involving powerful national actors. What is the foundation ofthis remark? I would argue the reverse. A lot of work exists on the role of rulesand institutions in networks.

7 That said, having the common values may facilitate making the decision.8 Of course, those same virtues about the public sector may make political change

more difficult and may require cultural change as well as (relatively) simple changesof structures and procedures.

References

Abbott, K. W. and Snidal, D. (2000) “Hard and Soft Law in International Govern-ance”, International Organization, 54: 421–56.

Bell, S. and Park, A. (2006) “The Problematic Metagovernance of Networks: WaterReform in New South Wales”, Journal of Public Policy, 26: 63–83.

Borras, S. and Jacobssen, K. (2004) “The Open Method of Coordination and NewPatterns of Governance in the EU”, Journal of European Public Policy, 11: 185–204.

Carpenter, D. P. (2001) Forging Bureaucratic Autonomy, Princeton, NJ; PrincetonUniversity Press.

Christensen, T. and Laegreid, P. (2007) New Public Management: Transformationof Ideas and Practice, Aldershot: Ashgate.

Dalton, R. J. and Wattenberg, B. (2000) Parties without Partisans, Oxford: OxfordUniversity Press.

DiIulio, J. J. (1994) Deregulating Government, Washington, DC: The BrookingsInstitution.

Elgie, R. (2001) Divided Government in Comparative Politics, Oxford: OxfordUniversity Press.

Hofstede, G. (2001) Culture’s Consequences: Comparing Values, Behaviors, Institutionsand Organizations across Nations, Thousand Oaks, Calif.: Sage.

Hood, C. (1976) The Tools of Government, Chatham, NJ: Chatham House.Hood, C. (1991) “A Public Management for All Seasons?”, Public Administration,

69 (1): 3–19.Jessop, B. (2002) “Governance and Meta-governance: On Reflexivity, Requisite

Variety and Requisite Irony”, in H. Bang (ed.) Governance, Governmentality andDemocracy, Manchester: Manchester University Press.

Lowi, T. J. (1972) “Four Systems of Policy: Politics and Choice”, Public Adminis-tration Review, 32: 314–25.

Mair, P. and Biezen, I. van (2001) “Party Membership in Twenty European Demo-cracies, 1980–2000”, Party Politics, 7: 5–21.

Milward, H. B. and Provan, K. G. (2000) Managing the Hollow State, Journal ofPublic Administration Research and Theory, 10: 359–80.

Moore, M. (1995) Creating Public Value: Strategic Management in Government,Cambridge, Mass.: Harvard University Press.

Mörth, U. (2004) Soft Law in Governance and Regulation, Cheltenham: Edward Elgar.Mueleman, L. (2008) Public Management and the Metagovernance of Hierarchies,

Networks and Markets, Heidelberg: Physica.Mulgan, G. (2000) “Accountability: An Ever-expanding Concept”, Public Administra-

tion, 78 (3): 555–73.Mulgan, G. (2009) “Accountability”, in IPSA Encyclopedia of Political Science,

Thousand Oaks, Calif.: Sage.Nicholson, N. (2002) “Policy Choices and the Use of State Power: The Work

of Theodore J. Lowi”, Policy Sciences, 35: 163–77.Ostrom, E. (1996) “Covenants, Collective Action and Common Pool Resources”, in

K. E. Soltan and S. L. Elkin (eds) The Constitution of Good Societies, UniversityPark, Pa: Penn State University Press.

Painter, M. A. and Peters, B. G. (2009) Administrative Traditions and AdministrativeChange, Basingstoke: Palgrave Macmillan.

Peters, B. G. (2001) The Future of Governing, 2nd edn, Lawrence, Kan.: UniversityPress of Kansas.

Salamon, L. M. (2001) “Introduction”, in L. M. Salamon (ed.) Handbook of PolicyInstruments, New York: Oxford University Press.

Savoie, D. J. (1994) Reagan, Thatcher, Mulroney: In Search of the New Bureaucracy,Pittsburgh, Pa.: University of Pittsburgh Press.

Scharpf, F. W. (1988) “The Joint Decision Trap: Lessons from European Union Integration and German Federalism”, Public Administration, 66 (2): 239–78.

Scharpf, F. W. (1994) “Games Real Actors Could Play: Positive and Negative Coordination in Embedded Negotiations”, Journal of Theoretical Politics, 6:27–53.

Scott, C. (2004) Regulation in the Age of Governance, Cheltenham: Edward Elgar.Scott, G. (1996) Government Reform in New Zealand, Washington, DC: Inter-

national Monetary Fund.Smismans, S. (2008) “The European Social Dialogue in the Shadow of Hierarchy”,

Journal of Public Policy, 28: 161–79.Sorenson, E. (2006) “Metagovernance: The Changing Role of Politicians in Pro-

cesses of Democratic Governance”, The American Review of Public Administration,36: 98–124.

Walsh, K. and J. Stewart (1992) “Change in the Management of Public Services”, Public Administration, 69: 499–518.

INNOVATIONS IN GOVERNANCE

Mark Moore and Jean Hartley

Introduction

This chapter explores a special class of innovations – innovations in govern-ance – and develops an analytical schema for characterizing and evaluatingthem. To date, the innovation literature has focused primarily on the pri-vate rather than on the public sector, and on innovations which improve organizational performance through product and process innovations ratherthan on public sector innovations which seek to improve social performancethrough reorganizations of cross-sector decision-making, financing and pro-duction systems. On the other hand, the governance literature has focusedon social coordination but has not drawn on the innovation literature. Thechapter uses four case studies illustratively to argue that innovations in governance deserve greater theoretical attention. Further, it argues that fiveinter-related characteristics distinguish public sector innovations in governancefrom private-sector product and process innovations. Innovations in govern-ance: go beyond organizational boundaries to create network-based financing,decision-making and production systems; tap new pools of resources;exploit government’s capacity to shape private rights and responsibilities; redistribute the right to define and judge value; and should be evaluated interms of the degree to which they promote justice and the development of asociety as well as by their efficiency and effectiveness in achieving collectivelyestablished goals.

Innovations in governance as an emerging research domain

Recently, there has been a great deal of both professional and scholarly inter-est in “innovation” in the public sector (Altshuler and Behn 1997; Borins1998; Hartley 2005; Moore 2005; Mulgan and Albury 2003; Albury 2005;National Audit Office 2006). Among professionals, innovation is seen as acritical method for improving the performance of government, and enhancinggovernment’s legitimacy with citizens. Nothing other than an innovative government can keep pace with citizen aspirations for a government that is

both efficient and effective in its core operations, and capable of respondingto the diverse and changing needs of a modern society.

For their part, public management scholars (Hartley 2005, 2006, 2008;Osborne and Brown 2005; Landau 1993; Walker et al. 2002; Koch andHauknes 2005; Moore 2005) have focused on the kinds of changes in govern-ment1 that should count as important innovations, how much innovationoccurs, what structures and processes promote or retard innovations, andwhether current institutional structures and processes support enough innovation for government to respond to or anticipate societal demands forchange.

Our focus in this chapter is on what we claim is a distinct class of publicsector innovations that deserve special attention: a class of innovations thatwe characterize as “innovations in governance”. These innovations differ fromstandard intraorganizational innovations in products, services and productionprocesses in at least two obvious ways. On the one hand, the innovationsare conceived and implemented above the organizational level: they involvenetworks of organizations, or the transformation of complex social produc-tion systems rather than changes solely within a particular organization. Onthe other hand, these innovations focus not only on concrete changes in what particular things are produced through what particular production processes, but also on the ways in which productive activity is financed (or,more broadly, resourced), the processes that are used to decide what will beproduced, and the normative standards used to evaluate the performance of thesocial production system.

We focus on this class of innovations for two different reasons. On the onehand, we argue that these innovations are theoretically interesting becausethey challenge conventional ideas about what constitutes an importantinnovation, where and how innovations are produced, and how they mightbest be evaluated. On the other hand, these innovations are important in practical terms. Those innovations in the public sector that are brought to public attention, or which win awards, often have the characteristics of being important or significant because they change where and how innovation is produced, as noted above. For both theoretical and prac-tical reasons, then, we suggest that this class of innovations deserves closeattention.

We want to be clear, however, that our approach to this subject is exploratory. We suggest but cannot confirm that we have identified a set ofpublic sector innovations that are both similar enough to one another anddifferent enough from other kinds of innovations to be considered as a coherent class of innovations. And our suggestion that, in practice, this classof innovations is both common and important among all public sector innovations requires further investigation. All that we are sure of is that thereare some kinds of innovations now being developed in the public sector thatdo not resemble the standard picture of product and process innovations

that occur in manufacturing and service-delivery firms in the private sector,or the standard picture of program innovations in public-sector organiza-tions. We suggest, however, that it is likely that in explaining how such inno-vations occur, how to evaluate their impact, or how to teach people how todo more of them or make them more successful, we shall have to go beyondexisting theories of innovation. This chapter is part of the journey of dis-covery about such innovations, their explanation and impact.

Evidence that “innovations in governance” have been an important partof the overall level of innovation in government is not hard to find. Mostdescriptions of important innovations in the public sector tend to focus on these kinds of innovations as well as on more traditional product andprocess innovations. Evidence that innovations in governance have given the academics trouble comes from the fact that when the theoreticians seek to categorize different kinds of innovations they often start with thefamiliar product, process and technology categories, but end up having to create some kind of residual category. Walker et al. (2002), for example, include the concept of “ancillary innovation”, defined as that which involves “organization–environment boundary innovations”. Mulgan and Albury(2003) talk about “systemic innovation”, which results from or is based onthe development of new underpinning technologies (or production systems)and/or organizational forms necessary to sustain and guide these new production systems. Only Hartley (2005, 2008) mentions the concept of governance innovations, in an outline of the dimensions (not categories) of innovation. These include changes to institutional forms of government(such as the devolution of power from national government to newly estab-lished governments for Wales and Scotland) and changes in organizationalform and arrangements for the planning and delivery of services (e.g. pri-vatization, new collaborative arrangements between the public and privatesectors) as well as those innovations that provide for greater public and/oruser participation in service design and delivery, and in the use of boards to govern particular choices and services (e.g. school governing bodies).

Importantly, when we turn to the literature for guidance about how tounderstand such innovations, we find a gap. The innovation literature (forboth private and public sectors) helps us understand what counts as an innovation. That literature makes it clear that innovations have to be morethan mere ideas: innovations are new ideas and practices brought into imple-mentation (e.g. Bessant 2005; Tidd et al. 2005; van de Ven 1986; Wolfe 1994).They are therefore different from inventions (Bessant 2003). Some commen-tators also add that an innovation is different from continuous improvementor other minor changes. For example, Lynn (1997) argues that “Innovationmust not simply be another name for change, or for improvement, or even fordoing something new lest almost anything qualify as innovation. Innovationis properly defined as an original, disruptive, and fundamental transformationof an organization’s core tasks” (p. 154). In this formulation, innovation

is a step change for the organization (see also Utterback 1996). Moore andcolleagues (1997) argue that innovation is “large enough, general enough and durable enough to appreciably affect the operations or character of theorganization”. Innovation may also include reinvention or adaptation of an innovation in another context, location or time period (Rogers 2003;Thompson 1965).

But all of this discussion of innovation focuses on innovations within organizations – important changes in what organizations produce and deliver.It does not focus on changes in structures and processes operating above the level of a single organization, and on financing, decision-making, andevaluations of performance as well. If, then, a public-sector innovator hadmade changes above the level of a given organization that transformed socialconditions by changing the social structures and processes that generatedresources to deal with the problem, or changed who was responsible for doingthe work, or allowed individuals to make voluntary contributions to the socialpurposes in exchange for more control over what was being produced withsocial resources, the existing literature on public sector innovation seems to have relatively little to offer. The public sector literature on innovationwill need to conceptualize and theorize innovations in governance as well asinnovations in government organizations.

When we turn to the governance literature, in contrast, we find a litera-ture that is attuned to an analysis of the broader social systems that guide,finance and produce large social outcomes. Kooiman (2003), for example,has defined governing as “The totality of interactions, in which public as wellas private actors participate, aimed at solving societal problems or creat-ing societal opportunities; attending to the institutions as contexts for thesegoverning interactions; and establishing a normative foundation for allthese activities” (p. 4). Government, as an important governance institution,is able to use its powers to convene actors from different sectors, and bothto regulate and to finance their activities might play an important role inintroducing innovations in the (social-level) governance of the social productionsystems. But this literature has not focused on how to analyze particularchanges made in governance processes, and their implications for the trans-formation of both social conditions and the redistribution of the burden ofproducing those social conditions, or on the methods that were used to bringthis new governance system into existence and to sustain its operations. Inshort, the governance literature misses the detailed operational focus thatcharacterizes the innovation literature.

The gap in the academic literature on innovations in governance, notedby Hartley (2005), is unfortunate, for the practical world seems to be producing a great many innovations of this type. Indeed, we seem to be going through a revolution in the governance of public production systemsas governments seek to reach beyond their borders to find additionalresources, additional operational capacity, and even additional legitimacy

to achieve their assigned goals. Some of the innovations involve changes in organizational-level governance – for example, a school is required to establish a parent’s council that can oversee its operations, or patients in public hospitals are given increased powers to voice their concerns about service quality. In other situations, the innovations involve new ways of knit-ting elements of different organizations together to create a more effective problem-solving approach to a given problem (Skelcher 2005). Hill and Lynn (2005) argue that “the focus of administrative practice is shifting from hierarchical government towards greater reliance on horizontal, hybridized,and associational forms of governance” (p. 173). These shifts, in line with other changes associated with “networked governance” (Benington 2000;Newman 2001), have implications for management, both in terms of organ-izational and inter-organizational processes, and potentially for performance.

Hierarchical government has been able to harness the use of state authorityas well as resources to achieve outcomes, sometimes coercively (e.g. throughlegislation about taxation and military powers) and through its claim to havea democratic mandate. The shift to achieving societal goals through partner-ships with the private, voluntary and community sectors means that influ-ence becomes a significant strategy as well as (sometimes instead of ) formal hierarchical authority (Hartley and Allison 2000). This has implications forthe ways in which managers undertake their tasks and for the organizations,partnerships and networks within which they do this. Hence innovations ingovernance become important to analyze as well as service innovations.

The gap in the literature is not hard to fathom. Both academics and pro-fessionals who have sought guidance about how to produce value-creatinginnovations have generally turned to the private sector for inspiration andguidance. They do so for two reasons. First, the private sector tends to giveinnovation a more prominent place in improving performance than govern-ment has done. Second, there is simply more scholarship on private than on public management. Consequently, in seeking scholarly guidance aboutthe role of innovation in improving governmental performance, and the processes that can foster value-creating innovation, one naturally turns tothe private sector literature. However, the private-sector literature has notexamined innovations in governance, and there are few studies which have examined innovation above the firm level (Hartley 2005). Many of the innovations which concern the public sector are not the process and product innovations that have been the meat and potatoes of innovation in the private sector. They are, instead, innovations designed to reshape a broader social system that not only produces public goods and services (and, in doing so, to transform aggregate social conditions in socially desirable ways), but also provides the financing and material to produce these results. They also shift the location of decision-making authority overthe new system to determine how the benefits and burdens of that new system will be distributed.

Method of inquiry

The purpose of this chapter is to initiate a more sustained, detailed inves-tigation into this particular class of innovation that seems to be both very important in government and less well understood by the private sector literature.2 The method is to look closely at a small number of public sector innovations (based on documentary evidence, obtained from teachingcases and from official reports and the media) that seem to fall within thebroad set of innovations in governance, taking care to ensure that the selec-tion of the cases includes some important variants. We do not offer thesecases as reliable histories of events, nor as complete evaluations of their social impact. Nor do we offer these cases as a representative sample of innova-tions. We offer them, instead, as particular instantiations of innovations ingovernment that do not seem to fit the accepted frame of product and pro-cess innovations discussed in relation to the private sector. They provide an opportunity to explore these apparently anomalous forms of innovation.The value of the cases lies in their ability to challenge our conceptual thought,not in their historical accuracy or representativeness. From the cases we developfive propositions about how innovations in governance are distinctive fromproduct and service innovations.

Innovations in governance: some illustrative and challenging examples

Contracting with community groups for child protection services

The Massachusetts Department of Social Services was experiencing consider-able difficulties in achieving the goals of their Child Protection Services (CPS)programme. Preventing abuse and neglect of children had to be addressedwithin financial constraints and with due regard for the privacy of familiesand the rights of parents. As the agency sought to balance the interests ofthe care of children, on the one hand, and the rights of families and parents,on the other, it made decisions with negative consequences of two kinds: failing to intervene where abuse or neglect was subsequently found, and intervening where the cases turned out not to require action. The difficultyof making the appropriate response in circumstances that are inherently complex, dynamic and unpredictable (cf. Hoggett 2006), together with thesensitive and ambivalent nature of the issues involved, meant that the CPSsuffered from a chronic threat to its legitimacy and effectiveness.

Such problems were particularly marked among immigrant communitiesof Boston. Many did not trust the intentions, methods or procedures of the CPS. They thought that the CPS did not adequately understand the culture of local communities, which affected what constituted good and badparenting in that context. They did not think the agency obtained accurate

information from individuals about family conditions or interpret it properly.They did not think the agency had much to offer them when there wereinstances of abuse and neglect.

Faced with this perceived crisis in the performance and legitimacy of theCPS, which was also in serious financial difficulty, the Massachusetts Depart-ment of Social Services developed an “innovative” approach which was basedon contracting out the service and which implicitly shifted the govern-ance arrangements through delegating the responsibility for receiving andresponding to complaints about child abuse and neglect to community-basedorganizations because they enjoyed much closer connections to, and muchgreater legitimacy with, immigrant communities. This seemed innovative not least because it tapped into a wider set of capacities and resources thanthe agency possessed: local knowledge of the customs and mores of parent-ing, ability to obtain and interpret information about conditions within a family, and a capacity to make interventions that would feel appropriateand useful to the affected.

However, behind this change lay some troubling questions for the observer.For example, was the state delegating either the de facto or de jure right todefine what constituted abuse and neglect to a community-based organiza-tion? If not, what decision-making and administrative systems would ensurethat the community-based organizations applied CPS standards accuratelyand consistently? What would happen to the legitimacy and effectiveness ofthe community-based organization if it was required to enforce CPS standardsof care and intervention? Would such a move undermine exactly the kind ofinnovativeness that the state was seeking through this contract?

To make the arrangement workable, both organizations had to learn torecognize their own and the other party’s interests and negotiate differently.The government agency had begun with the goal of contracting out childprotection services, defined largely in terms of hearing and responding to com-plaints about child abuse and neglect. The community-based organizationdefined its interest in persuading the CPS to give it money for the provisionof services to clients, without taking responsibility for setting and enfor-cing standards. In fact, these views dominated the actual negotiations and led to a contract that was somewhat cynical on both sides: the CPS claimedto have widened responsibility for abuse and neglect services (while actuallyshifting its responsibilities and lowering its costs), and the community-based organization accepted this responsibility but without really taking the full responsibility for doing the work or accurately pricing the level and activities required. The inadequacy of the contract was exposed later when a child in the care of the community-based organization was found to have been seriously abused. A formal investigation showed that the community-based organization had not, in fact, taken the kind of con-sistent responsibility for the care of children that the CPS claimed to havecontracted for.

This case illustrates that innovation does not necessarily lead to improve-ment (see also Hartley 2005 for this distinction). There was potential for aninnovation which would have enhanced public value (Moore 1995; Beningtonand Moore forthcoming) by paying explicit attention to the governance as well as the service innovation. Such an innovation would need the community-based organization to organize a community-based discussionabout problems of abuse and neglect and ways to address child protec-tion. It might have orchestrated such a discussion either on its own, or withthe CPS and possibly other interested stakeholders. Then, following that community-based discussion, a whole system of prevention and interventioninvolving individual and collective, community-based and governmentalaction might have been developed. The difficulty for the CPS was that it had been unable to focus on this as an innovation in governance arrange-ments because they would only contract for service or process innovationsin child protection.

Private partnerships to support New York City’s parks

The New York park system was once one of the glories of life in New York.Initially conceived in the mid-nineteenth century by Frederick Law Olmstedas oases from urban squalor, and then extended as a wide network of easilyaccessible green spaces, New York’s parks had long been a refuge for city-dwellers. By the late 1960s, however, the parks were falling into disrepair.The gardens were trampled; trees were vandalized; the greenswards were dustyand littered; the recreational equipment was broken. They were less oftenand less widely used because they seemed, and actually were, increasinglydangerous.

The city government parks organization had become overwhelmed, withinsufficient financial and staff resources to run the city’s parks. It could notrely on citizens to use the parks well nor could it generate public commit-ment to the parks. Senior managers decided on a new approach. Instead of the organization acting as though it was the only body responsible forthe parks, they decided to reach out for partnerships with citizen groups, toencourage a greater interest in the parks in exchange for somewhat greatercontrol over what happened within them. The partnerships they created took different forms in different parks, but in each case citizen groups were invited to contribute direct resources to the park. This was not throughtaxation (which ensures that the costs of park maintenance are fairly dis-tributed amongst all citizens, but which reduces the perception of a person-alized effect of contributions). Instead voluntary contributions of time andmoney to particular parts of, or particular activities within, each park were solicited. From one perspective, such partnerships might be viewed as“selling” a piece of the public park to a particular set of users in exchangefor an additional voluntary contribution of labor or money. In practice, the

newly renovated parks and the new, jointly sponsored activities were not exclusively for the contributors; because they were (at least in principle) stillavailable to all. The volunteers may have felt particularly attached to thepark renovation, and they may have felt some special entitlements to use the space and to host those who came to use the parks. But the parks retainedtheir public character in that they were free to all and the overall set of usesfor the parks did not change.

Through these partnerships, the New York City parks bloomed again. Theybecame prettier, safer, and much more widely used without costs to governmentincreasing. Arguably, the public value of the parks had been enhanced. Onthe other hand, the commitment of voluntary time and resources created acertain degree of informal moral agency and claim over influencing the debatesover public purpose, as we shall explore later.

Congestion charging in London

London is widely viewed as a world city, the powerhouse of the British economy, and an international gateway for investment and tourism. Yet it has been dogged by an inadequate transport system, which is seen as limiting economic growth and the quality of life of its citizens, workers andtourists. A combination of problems (underinvestment in public transport,deterioration of the railway system following privatization, fragmenta-tion of decision-making about infrastructure including transport planningand provision) had left London at the turn of millennium with major traffic-congestion problems. Private and commercial vehicle use in central Londonhad become slow and unpredictable, affecting business and leisure time. Publictransport was unreliable, giving car-drivers little incentive to use public trans-port instead.

The development of an innovative solution, partly in the form of con-gestion charging, came from a particular combination of circumstances. TheTransport Strategy was developed in recognition by politicians and managersthat four factors coincided in a way which meant that it was possible to startto resolve London’s transport crisis. First was the innovation in governanceof the establishment of devolved government for London as a city. The newGreater London Authority, with a directly elected mayor with a manifestoabout improving transport and travel, gave a strong democratic mandate totackle transport problems. In addition, at the same time, London’s trans-port services were integrated through the establishment of a new organiza-tion, Transport for London (a strategic innovation which underpinned partof the governance innovation). Third, the central government, still newlyelected, provided a level of financial resources which helped to tacklechronic underinvestment. The fourth element was the hiring of key seniorinternational managers with a proven track record of tackling transport problems. Each of these elements can be considered an innovation in its own

right, and they were used together to develop an innovative strategy for redu-cing congestion in central London. A charge was introduced, from February2003, for using a vehicle (other than taxis) in the central eight square milesof London during the day. This was the first time for generations that roadsin London had been subject to a toll for use.

Initially, this approach faced a number of political and technical/operationalproblems. Establishing a consistent and fair way of warning travellers of imminent entry into the charging area, and monitoring road use so that the charge could be applied were important. So was ensuring that pay-ment and enforcement was effective, efficient and feasible, with travellers having access to information about other forms of travel. There was also a challenge to ensure longer-term viability of the scheme, and to encouragebehavioral changes in the travel habits of the millions who lived and workedin London.

The political challenge was to create a vision and mobilize for the pro-posed changes, with long-term commitment to the innovation. When the policy was first proposed by the mayor, chaos and disaster were predicted bythe opponents of the scheme, and even ordinary Londoners were scepticalabout whether it would work. Civil disobedience, traffic gridlock in the areajust outside the charge zone, and intolerable pressure on the bus, Under-ground and rail networks were all predicted. The plans of the elected mayorand the newly devolved Greater London Authority, along with Transportfor London, were all put under the spotlight by the media and by lobbyinggroups, though some groups were supportive. Politicians took time and careto outline London’s problems, to explore options and to listen to concernsabout the new scheme. Managers held consultation events around Londonto learn about the ways in which different groups might be affected. An import-ant issue was to ensure that congestion charging shaped travel behavior in fundamental ways, not merely raised funds for the city. In other words,citizens and visitors had to learn to adapt to the new system. Since the intro-duction of charging, car traffic has reduced by about 30 percent, businesshas benefited from shorter and more reliable journey times, public transporthas (largely) coped, and cycling has increased.

Elder care in Singapore

Singapore’s highly centralized national government is committed to, and has achieved, dramatic economic growth fueled by foreign direct investment.Among the conditions that attracted investment was a government that pro-tected private property rights and that could promise labor peace. Partly toachieve this, government guaranteed access to high-quality housing, whichwas highly prized by families.

Economic development gradually produced strains on the social and governmental structures. There was an increasing demand for governmental

processes that were more open and democratic – that allowed or encourageddebates about public policy, and made governmental actions more account-able. Also, as society became exposed to more individualistic Western cultures, social relationships in families were altered, and concern arose aboutan aging population. Longevity was increasing; and, despite the decades ofeconomic growth, it was possible that those who had created that growthwould not be financially secure in retirement.

To many, the aging population was not a problem because a well-established social custom located responsibility not on the individual or the state but on offspring. They had a duty to attend to their parents’ needswhich had been strong enough to produce both reliable care from the vastmajority of Singaporeans and vigorous informal criticism of anyone whoseemed to abandon their aging parents.

However, the customary system had always been incomplete (for example,what about elders with no children?). It had always been imperfectlyenforced (there were some children who neglected their parents, and thoseparents had no formal right of action against their children). In addition,there was concern in government that customary duties were weakening under the influence of both Western ideas and a sense that the state wouldprovide. It was conceivable that this customary system would break down,leaving many elderly people exposed to penury and loneliness.

This issue was taken up by an appointed legislator as his particular cause. He proposed a new public law to underpin the customary duty. Thelaw required children to care for their parents in kind or by financial contribution, and gave neglected parents recourse to the law. It also allowedthe state to pursue children who failed in their duty of care. The shift from the customary system to backing with statutory requirement and a right of action was a significant innovation.

Preliminary considerations

We think it is fair to say that these sorts of innovations differ from the product/service/process innovations that have been the focus of such exten-sive attention in the private sector literature. But many of the most widelyremarked and celebrated innovations in the government sector seem to beof these broader, more structural types where production, financing and decision-making are all moved around in a new configuration to reshape the system that determines what is produced, how it is financed, and whosevalues are given emphasis in guiding the process of social production.

How are we to understand these innovations? They seem to work (to varying degrees) in practice, but where do they fit in our theories of inno-vation? How might the analytic frameworks we use for characterizing andevaluating innovations have to be changed to accommodate these broader,more structural types?

Are innovations in governance really innovations?

Let us start by asking whether changes like the ones above deserve to becalled innovations and, if so, why. We can then turn to the question of what,if anything, makes them different from product/process innovations.

The innovations described above may logically entail or create the condi-tions under which many different process and product innovations can occur. For example, it is quite likely that the new governance arrangementsin New York’s Central Park will generate a wider variety of uses of the park ranging from gardening, to bird-watching, to ethnic festivals. Further,for each of these new uses (or services) a different production and financingsystem might be generated.

Similarly, the new system for governing the rationing of the roads in Londonmight require the development of many new products and activities that per-mit the charging of individuals for travel – the technical arrangements thatallow us to make what was once a freely used resource one where use is moreexclusive through noting who is using the product/service and charging them for it. But, while each of the governance innovations has dimensionsof production and service innovation, that is not the whole story.

One can also raise doubts about the degree to which these ideas are genuinely new. The fact that these innovations seem to reach out to privateassociations and private individuals to accomplish public purposes does not seem particularly new. Society, acting with or without the help of government as its agent, has always relied on or been shaped by charity and civic action with or without the financial encouragement and directionof government. Similarly, we have long been accustomed to the idea thatprices can be used not only to raise revenues for the seller, and to divide thevalue of creating a product or service that is desired by a customer betweenthe producer and the user of that product and service, but also to ration limited supplies of a given product, and to channel the products and services to those who want it most (conditional on their ability to pay). Wehave used this idea not only in the private marketplace, but also in manag-ing the level and distribution of production for such utilities as water, electricity and communications. So it does not seem such a big innovationto use it as a device for rationing road use. And we have long understoodthat public purposes such as elder care could be advanced by requiring indi-viduals to act in accord with public laws as well as by relying on existingmoral commitments to induce individuals.

Still, what makes the cases interesting as innovations is that they do, infact, change the location and financing of social production, and the leveland distribution of things that could reasonably be called social or publicgoods and services. The level, character and distribution of child protectionservices change as community-based groups are drawn into the process with government authorization and contracts. The fact that they are drawn

into the production process gives them at least de facto and perhaps de jureroles in deciding what will be produced, for whom, and in what ways, with important consequences for both the parents (whose conduct is now monitored differently) and the children (whose welfare depends so much on the actions of parents). The level, character and distribution of park services change as the new partnerships are initiated and sustained – and with those changes an alteration in the observed character and utilization of the New York City park system. The level, character and distribution ofthe London roads service changes when congestion charging is introduced.The level, character and distribution of aid to aging parents is altered whenlegislation imposing this duty on children is discussed, passed and enforcedthrough private and public means. And so on.

It is because these innovations change what is produced, how the new pro-ducts and services are distributed, how the burden of producing the servicesis borne, and what happens to the material conditions in society that these“innovations in governance” deserve to be taken seriously as innovations.If they did not produce these material changes in what is produced for whom,and how the aggregate social conditions are changed as a consequence, thenthey would not be interesting as an important class of social innovations.

Five ways in which these innovations are different

The fact that these self-consciously constructed and introduced measures changethe material processes through which society seeks to deal with particularproblems makes them innovations that are worth noting as innovations. Fromthe point of innovation theory, however, what makes them particularly inter-esting is all the ways in which they are not like the innovations in productsand services. They seem to differ in at least five, highly inter-related ways.

Bursting the boundary of organizations/creating network-based production systems

First, the innovations described above seem to burst the boundaries of any particular organization, and to relocate and redistribute where and how socially productive activity occurs. The contracts with community-based organizations shift the production of child protective services from a statebureaucracy to a network of community-based groups. The invitation to private agencies to contribute their efforts to the maintenance of the parksshifts both the production and use patterns of the parks from one that wasset by the Parks Department to one that is set by the Parks Department working in a network of partnerships. London’s congestion pricing systeminvites drivers in London to find other means for meeting the objectives theypursue by using London’s streets.

In each of these innovations, a particular organization stops being the sole locus of change. Further, the organization’s future success stops beingthe sole focus of evaluation. Instead the focus of attention shifts from theanalysis of what happens inside an organization to an analysis of a pro-duction system that crosses organizational boundaries, and sometimes (as in the case of both congestion pricing and the law mandating the care ofaging parents) reaches to the mobilization of millions of decentralized indi-viduals. The way in which the innovation is evaluated, then, is not in termsof whether it increases the productivity or success of a given organization,but of whether it succeeds in altering the broad social conditions that havebecome the focus of some collective concern. That collective concern couldhave previously been seen as the exclusive responsibility of a given govern-mental organization, but has now been transformed by the innovation intoa problem to be solved by a much wider production system that stretcheswell beyond the resources that can be directly controlled by any given government organization.

Indeed, it is precisely this move to burst the boundary of an organization’shold on a given (and complex) problem that represents an important part of the innovation. As long as a given problem was held within a givenorganization, and as long as society relied on that bounded organization to solve the problem, the problem could not be fully addressed. It was onlywhen the society, acting through the agency of government, decided toinvite other actors into the solution of the problem that an importantchange could be made. These innovations are less organizational innova-tions, then, than system innovations that reconfigure production systems forachieving a given social result.

Tapping new pools of financing, material resources and human energy

Second, in many cases, innovations in governance focus not only on chan-ging production systems, but also on tapping new wellsprings of resources.Those new resources that are tapped can come in quite different forms. Some of the new resources involve specific bits of specialized operational capability that turn out to be valuable in achieving a particular purpose thegovernment has in mind. In the case of the community partnerships for Child Protective Services, the State Agency hoped to tap into an asset thata community program has naturally, and has further developed over time:namely, its established knowledge of, and legitimacy with, the local community.In the case of elder care in Singapore, the innovation is to strengthen a voluntarily contributed private capacity to care for the elderly with a legalobligation that will, ideally, add force and consistency to a voluntary customary practice.

At other times, the new resources come in a more fungible form; namelyfinancial contributions. In the case of the New York City parks, for example,an important part of the innovation seems to be allowing relatively wealthyNew Yorkers who want their parks to be nice to make voluntary contribu-tions of money. (The donors can make their contributions a bit less fungiblethan they first appear by conditioning their availability on an agreement thatthe government will use them in a particular way. But the specialization inthe use of the resources comes via institutional agreements rather than asmaterial aspects of the resources.)

Regardless of whether the resources come in the form of money, labor or material, and regardless of whether the resources are highly fungible orare specialized to some very specific purposes, one way that these inno-vations seem to work in helping to solve public problems is by locating and mobilizing resources that were previously on the sideline or not fullyexploited in the public effort.

Exploiting government’s capacity to convene, exhort, and redefine private rights and responsibilities

Third, in seeking to mobilize more heavily resourced and more effective pro-duction systems than it could when it was operating only through existinggovernment organizations with existing governmental resources, governmentrelies on different instruments to accomplish its ends. In the classic form of government-led public problem-solving, government assumes the fullresponsibility for defining a public purpose, mobilizing resources to solve it, and deploying those resources in the most efficient and effective way througha government agency. The principal operating instrument of the governmentis the taxes used to sustain the operations of a government bureaucracy. In the innovations described above, government uses different instrumentsto achieve its results.

In the case of the CPS, it uses finances not only to support a governmentagency, but also to contract with a private organization. It does so partlybecause the organization already has some capacity that the government needs and cannot easily develop, and partly because it might be able to usemoral suasion and the felt responsibility of the community group to makea greater contribution than it could buy from more professionalized or morecommercial enterprises.

In the case of the parental support bill in Singapore, government uses state authority to compel those who might be tempted to stray from theircustomary duty to their parents, and gives vulnerable parents a right to action against neglectful children. The creation of such an obligation has to be accompanied by sufficient resources to ensure that cases brought byparents can be heard in state courts. But the principal asset of the state thatis engaged is its authority to direct private action, and to mobilize the forces

of informal social control to help enforce the obligation, a force that mightbe strengthened or weakened through the passage of the law.

In the case of the New York City parks, the government attracts pri-marily money and some voluntary labor to improve conditions in the parks,and it does so by allowing private parties to make the contributions theywish to make and to earmark their funds for those purposes and places. Theimportant and interesting change here is that the Parks Department givesup its reliance solely on tax revenues in preference for accepting voluntarycontributions to the parks and, in doing so, gives up its exclusive power todecide how the public parks will be maintained and used. As the price ofaccepting voluntary contributions, government must negotiate with privateparties, and accept their ideas of what particular things they would like todo with the parks, as well as make decisions on their own about what thebest or fairest use of the park resources would be.

In these innovations, then, government not only uses its money to animateand direct activity of its own employees or contractors but also uses its directregulatory authority and its hortatory moral power to mobilize privateactors to make contributions to public purposes. It also allows individualsto make contributions to what were previously wholly government-controlledoperations and, in doing so, allows the contributors to begin to make changesto the results of the public system.

Redistributing the right to define and judge the value of what is being produced

Fourth, the innovations described above seem to change the locus of “decision rights” over the use of particular assets in society. This seems to come as an almost inevitable consequence of changing organizational boundaries and reaching out for private resources. When the state recruitsprivate money and community organizations to its purposes, it seems to giveup at least some of its power to define what should be produced, for whom,and in what way. The New York City Parks Department loses some of itsiron control over what happens in the parks. The CPS loses its iron con-trol over what happens in the handling of instances of abuse and neglect.Because it seeks some voluntary help in both cases, those who provide thehelp can negotiate the terms under which their help is offered. Because theyhave the power to “exit”, their “voice” becomes more powerful in shapinggovernmental policy and action. They do not have to remain “loyal” to thegovernment and its purposes.

On the other hand, the locus of decision-making and judgments about valuehave shifted in emphasis away from the individual to the state in relation tochoices over free access (London) and the duty of care to elderly people(Singapore). Decision rights that used to be held by individuals in a privatedomain had been powerfully reconditioned by government authority.

Evaluating the innovations in terms of justice, fairness andcommunity-building as well as of efficiency and effectiveness

Fifth, because these innovations use government authority as well as govern-ment money, and because they redistribute decision rights over the use ofboth publicly owned and privately owned assets, they invoke a different normative framework for evaluating the innovations we observe. In the classic case of private-sector product/process innovations, the innovations areevaluated largely in instrumental, utilitarian terms. The important questionsare whether the new production processes resulted in lower costs, or higherquantity or quality per unit of cost; whether the new product or service positioned an organization more effectively in its preferred markets; and there-fore whether it increased the prospects for maximizing shareholder wealth(as revealed in increased public valuations of its stock price).

In the cases considered here, where the innovations seem to relocate eitherresponsibilities for producing publicly valued results, or rights to decide whatconstitutes publicly valued results, or some combination of the two, one isforced, we think, to evaluate the innovations not only in terms of efficiencyand cost-effectiveness, but also in terms of what might be considered rightrelationships in the society – some notion of justice and fairness. After all,when a collective policy-decision is taken to move some established respon-sibility from the private domain to the public domain – as occurred whenthe Greater London Authority assumed the right to charge drivers for using certain London streets, or when the Singapore government legislated the obligation to provide elder care to the children – we are as interested in the question of whether that is a just and fair allocation of responsibility in the society as we are in the question of whether it will work to transformmaterial conditions in desired directions. Conversely, when a collective public policy-decision is taken to give private parties more power in shap-ing what were previously governmentally dominated operations – when, forexample, the CPS decides to give community-based organizations increasedrights to shape the local response to child abuse and neglect, or the NewYork City Parks Department allows private groups the right to make financial and labor contributions to the parks that are conditioned on theirparticular ideas of what would be a good use of that public asset – we arealso motivated to ask whether such a move is proper or not, and what the implications will be for the overall fairness and justice of a particularpublic production system.

Conclusion: innovations in governance as a challenge to innovation theory

In these five respects, then, the innovations in governance seem quite differ-ent from the innovations in products, services and production processes that

we have, until recently, associated with innovation in the private sector. Theseinnovations change production systems that cut across the boundaries of organ-izations, not just those of a single organization. They enlarge the range ofresources that can be tapped to enlarge and improve the performance of theproduction system. They involve changes in what instruments governmentuses to animate and direct the production system for achieving the desiredgoals. They alter the configuration of decision-making rights with respect to how private and public resources will be used. And they raise importantquestions about the distribution of burdens and privileges in the society.Precisely because they involve changes such as these, it does not seem unreasonable to describe these as innovations in the governance of societyand social conditions, not simply as innovations in government operations.

In considering the future of innovations in the public sector, innovationsin governance are a significant part. It is possible that innovations will con-tinue to evolve in ways which go to the heart of democratic government –the processes by which a community discovers its own interests, and beginsto speak coherently as a collective about its aspirations of justice, prosperity,social relations and ecological sustainability. John Dewey (1927) wrote inThe Public and Its Problems that the most important problem facing the public is discovering itself and identifying its own true interests. We arguethat this challenge will only be solved by more practice with, and innovationsin, the processes of democratic deliberation itself.

Acknowledgment

This is a revised and developed version of Mark Moore and Jean Hartley(2008) “Innovations in Governance”, Public Management Review, 10 (1): 3–20.

Notes1 In this chapter, we use “government” to refer to both government organizations

(e.g. federal and national government, local government, etc.) and public service organizations which may have a degree of autonomy from central government, such as health services, criminal justice services, and agencies concerned with theenvironment, public health, etc., but which are funded and regulated as part ofthe public service sector.

2 Note: there is an equivalent issue in the private sector: namely, when private firmsconstruct new contractual relations or, more ambitiously, new governance relationsto improve their individual firm performance (see Tidd et al. 2005). This includesmergers and acquisitions, which are evaluated in terms of the impact they have on the market position of the firms involved in the mergers. It also includes thecomplex bundles of ownership rights and responsibilities that have integrated high-tech bio-med firms. It may even include choices that socially conscious enterprisesmake about whether and how to form working partnerships with nonprofit organizations. For our purposes here, however, we will focus most attention onthese activities in the public sector where government is one of the important actorsin creating or acting within a particular governance scheme.

References

Albury, D. (2005) “Fostering Innovation in Public Services”, Public Money andManagement, 25 (1): 51–6.

Altshuler, A. and Behn, R. (eds) (1997) Innovations in American Government, Wash-ington, DC: The Brookings Institution.

Benington, J. (2000) “The Modernization and Improvement of Government and PublicServices”, Public Money and Management, 20 (2): 3–8.

Benington, J. and Moore, M. (forthcoming) In Search of Public Value, Basingstoke:Palgrave Macmillan.

Bessant, J. (2003) High-involvement Management: Building and Sustaining Competi-tive Advantage through Continuous Change, Chichester: John Wiley.

Bessant, J. (2005) “Enabling Continuous and Discontinuous Innovation: Learningfrom the Private Sector”, Public Money and Management, 25 (1): 35–42.

Borins, S. (1998) Innovating with Integrity, Washington, DC: Georgetown UniversityPress.

Dewey, J. (1927) The Public and Its Problems, Athens, Ohio: Swallow Press/OhioUniversity; reprinted 1988.

Hartley, J. (2005) “Innovation in Governance and Public Services: Past andPresent”, Public Money and Management, 25 (1): 27–34.

Hartley, J. (2006) Innovation and Its Contribution to Improvement: A LiteratureReview for Policy-makers, Policy Advisors, Managers and Academics, London:Department of Communities and Local Government.

Hartley, J. (2008) “The Innovation Landscape for Public Service Organizations”, in J. Hartley, C. Donaldson, C. Skelcher and M. Wallace (eds) Managing to ImprovePublic Services, Cambridge: Cambridge University Press.

Hartley, J. and Allison, M. (2000) “The Role of Leadership in Modernisation andImprovement of Public Services”, Public Money and Management, 20 (2): 35–40.

Hill, C. and Lynn, L. (2005) “Is Hierarchical Governance in Decline? Evidence from Empirical Research”, Journal of Public Administration Research and Theory,15 (2): 173– 95.

Hoggett, P. (2006) “Conflict, Ambivalence and the Contested Purpose of PublicOrganizations”, Human Relations, 59 (2): 175–94.

Koch, P. and Hauknes, J. (2005) Innovation in the Public Sector, Oslo: Publin.Kooiman, J. (2003) Governing as Governance, London: Sage.Landau, J. (1993) “Organizational Change and Barriers to Innovation: A Case

Study in the Italian Public Sector”, Human Relations, 46 (12): 1411–29.Lynn, L. (1997) “Innovation and the Public Interest: Insights from the Private

Sector”, in A. Altshuler and R. Behn (eds) Innovation in American Government,Washington, DC: The Brookings Institution.

Moore, M. (1995) “Creating Public Value”, Cambridge, MA: Harvard University Press.

Moore, M. H. (2005) “Break-through Innovations and Continuous Improvement:Two Different Models of Innovative Processes in the Public Sector”, Public Moneyand Management, 25 (1): 43–50.

Moore, M. H., Sparrow, M. and Spelman, W. (1997) “Innovation in Policing: FromProduction Line to Jobs Shops”, in A. Altshuler and R. Behn (eds) Innovation inAmerican Government, Washington, DC: The Brookings Institution.

Mulgan, G. and Albury, D. (2003) Innovations in the Public Sector, London: CabinetOffice.

National Audit Office (2006) Achieving Innovation in Central Government Organiza-tions, London: The Stationery Office.

Newman, J. (2001) Modernising Governance: New Labour, Policy and Society, London:Sage.

Osborne, S. and Brown, K. (2005) Managing Change and Innovation in Public ServiceOrganizations, London: Routledge.

Rogers, E. (2003) Diffusion of Innovations, 5th edn, New York: Free Press.Skelcher, C. (in press) “Does Governance Perform? Concepts, Evidence, Causalities,

and Research Strategies”, in J. Hartley, C. Donaldson, C. Skelcher and M. Wallace(eds) Managing to Improve Public Services, Cambridge: Cambridge University Press.

Thompson, V. (1965) “Bureaucracy and Innovation”, Administrative Science Quarterly,(10): 1–20.

Tidd, J., Bessant, J. and Pavitt, K. (2005) Managing Innovation: Integrating Techno-logical, Market and Organizational Change, 3rd edn, Chichester: John Wiley.

Utterback, J. (1996) Mastering the Dynamics of Innovation, Cambridge, Mass.: HarvardBusiness School Press.

Van de Ven, A. (1986) “Central Problems in the Management of Innovation”, Manage-ment Science, 32 (5): 590–607.

Walker, R. Jeanes, E. and Rowlands, R. (2002) “Measuring Innovation: Applyingthe Literature-based Innovation Output Indicator to Public Services”, PublicAdministration, 80: 201–14.

Wolfe, R. (1994) “Organizational Innovation: Review, Critique, and SuggestedResearch Directions”, Journal of Management Studies, 31 (3): 405–31.

GOVERNANCE ANDGOVERNABILITY

Jan Kooiman

1. Introduction

This chapter is not about governance in general. There are several introductionsavailable giving a fair picture of the many approaches to the concept (e.g.Schuppert 2005; Kjær 2004; Pierre 2000). Three features are common to them:they reflect the growth of social, economic and political interdependencies;governance is a matter of public as well as private actors; and dividing linesbetween public and private sectors become blurred. They differ mainly bydirecting themselves at a particular level, such as local, European or globalgovernance, or by focusing on a particular form or aspect, such as net-work, multi-level, or participatory governance. The governance perspectivediscussed in this chapter fits more in the second group looking at governanceas a societal phenomenon to be studied at all levels. In this sense Osborneis correct in saying that it can be considered as “most expansively . . . social-political governance as an over-arching theory of institutional relationshipswithin society” (2006: 381).

In its theoretical framework, interactions are given a central place as ittakes as its (normative) starting point that the solving of major problemsand the creation of major opportunities in modern societies are a combinedresponsibility of state, market and civil society together – be it in differentand shifting combinations of interactions between actors and institutions withinand between them (Kooiman 1993, 2003).

The more detailed elaboration of the governance approach now becomespart of the broader governability concept consisting of a system-to-be-governed (SG), a governing system (GS) and the interactions between thetwo (GI). The first part of the chapter is devoted to the interactive govern-ance perspective as such, and the governability framework forms the secondpart of this chapter (Kooiman 2008).

2. Interactive governance

In accordance with other approaches, the interactive governance perspectiveproceeds from the assumption that societies are governed by a combinationof governing efforts (Kooiman 2003). These governing mixes are “answers”to ever-growing societal diversity, dynamics and complexity, and responsesto major societal issues such as poverty and climate change.

The main concept here is that of “interactive governance”, defined as thewhole of interactions taken to solve societal problems and to create societalopportunities; including the formulation and application of principles guidingthose interactions and care for institutions that enable and control them.

The emphasis on “interactions” constitutes the main innovation in thisapproach. Interactions are specific forms of action, undertaken in order to remove obstacles and to follow new paths, whereby the definition of aproblem or an opportunity depends on the issue at hand as well as on theposition and understanding of the observer. The adjective “societal” is best-understood by way of its antonym, “private”, and is often replaced by theword “public” – it is everything that has a common, social and collectivecomponent. Institutions are also included in the definition as they are considered to be vital for any governance interaction. So, too, are principlesaccording to which interactions take place and institutions function. Theassumption is that governance arrangements lacking a normative basis sufferfrom ineffectiveness and illegitimacy in the long run.

Theoretically the interactive perspective on governance proposes thatsocieties are made up of large numbers of governance actors, who are con-strained or enabled in their actions by structures. Actors, in this perspective,are any social unit possessing agency or power of action. These include indi-viduals, associations, leaders, firms, departments and international bodies.Structure refers to the frameworks within which these actors operate; theselimit or widen their action potentials, and must therefore be taken into account.These frameworks include culture, laws, agreements, material and technicalpossibilities. According to sociological reasoning, actors are continuously making changes to these structures while at the same time being subjected totheir influence (Giddens 1984; Berger and Luckmann 1966). The analysis ofgovernance requires attention to both dimensions.

Present-day societies derive their strength from their diversity, complexityand dynamics.

Diversity calls attention to the specific and varying qualities of actors andother entities in a societal system. It is a source of creation and innovation,but also carries the danger of disintegration. Complexity invites examin-ation of societal structures, interdependencies and inter-relations, and is a condition for combining interdependencies. The difficulty is how to reduceit in an effective and responsible manner. By introducing dynamics of sys-tems we call attention to the regularity or irregularity with which societal

GOVERNANCE AND GOVERNABILITY

developments, often combined with tensions, take place. Dynamics createthe potential for change, but can have disruptive consequences.

These features continuously present societies with problems, but also withopportunities. These opportunities and problems themselves are also com-plex, dynamic and diverse. After all, they reflect the strengths and weaknessesof these societies. This also applies to the institutional conditions under whichopportunities are created and seized, and problems formulated and solved.To satisfy governance standards, such as being effective, efficient, legitimateand social-political or interactive, governance itself has to reflect the diverse,dynamic and complex character of the challenges it faces. Often problemdefinitions are too simple, policies too static, and audiences too generalized:this might be one of the primary reasons why so much governance does notsatisfy governance criteria as stated.

3. Governability

Governors, the governed and the nature of interactions among governorsand the governed all contribute to the governability of societies or parts thereof.In other words: governability is not considered as something primarily dueto the quality of those governing or governance entities – as in older con-cepts of governability often was argued (for an overview of this aspect, seeKooiman 2008); in the interactive governance perspective, governability isconsidered to be a quality of societal systems as wholes. For that reason, itcan be defined as the overall capacity for governance of any societal entity orsystem as a whole.

The interactive governance approach assumes that the condition of govern-ability of any system is continuously changing in response to external andinternal challenges. What may be high governability at a given time may below governability at another. Similarly, what may be effective governancein one place may be ineffective in another. Acts of governance may influencegovernability as a whole or any of its components. However, many externalfactors influence governability as well, some of which cannot, or only incom-pletely so, be handled neither by those governing or those governed at the scalethis is applied. This often enhances uncertainty with respect to the govern-ability of a societal system or entity in its human as well as its natural parts.One may state that governability is a condition fluctuating on a continuingscale without ever reaching the extremes; it is never total or complete.

Systems

Systems are usually defined as the whole of inter-relations among a givennumber of entities. Societal systems are “rich” in the sense that their com-ponents and their inter-relations have many facets, will overlap in many ways,all have their own histories and identities. Systems show properties which

cannot be derived from simply adding properties of the constituting com-ponents. What a system looks like, how it can be broken down, and whatits boundaries and other qualities might be depend on the perspectives of itsobservers. Any system – societal, natural, or combinations of the two – canbe seen as part of a set or hierarchy of nested systems. Where in this set onewants to locate a particular system also exists in the eyes of its beholders.The more beholders with comparable ideas about a system, the stronger theconcept becomes, for study and for practical purposes. The flexibility of definingsystems, their parts, their boundaries and environments for particularresearch purposes is one of its attractions. This general observation also appliesto societal systems.

The diversity, complexity and dynamics of the inter-relations between partscharacteristic for systems (societal and natural) make for uncertainties andunpredictability in systems behavior. There is nothing secretive about thesequalities; they simply are the consequence of actors, elements or parts of systems acting or interacting without having the possibility of knowing whatthe results of their actions or interactions are for systems behavior as a whole.Systems theorists such as Jervis (1997), Gell-Mann (1994) and Cilliers (1998)all put great emphasis on this aspect.

Mechanisms like these have consequences for those (inter)acting in a societal system, but also for the study of them, because basically reduc-tionist approaches focusing on actors or interactions by themselves must failbecause the larger (system) picture misses (see, e.g., Mayntz 2004). But a purelyholistic one, where the system as such is the unit of analysis, does not work,either. Combinations are necessary. By stressing the role of interactions aswe do, one step toward a better understanding of governance behavior canbe set.

Important is the conceptual relation between scale, level, hierarchy andnestedness as we see the analysis of governance and governability of aspecific system as a choice of level in a set of nested systems. Scale is seenas the spatial, temporal, quantitative, or analytical dimensions used to measureor study any phenomenon and level as a unit of analysis located at the same position on a scale. Scales and levels can serve different analytical ortheoretical purposes.

To understand what a system might look like, it must have boundaries inone way or another. Societal systems with natural components or resourcesincluded are open systems – up to a point. This is of importance not onlytheoretically but also at a fundamental level. Completely closed systems, according to the laws of entropy, will die in the long run, because every system needs outside energy and/or information to survive. On the other hand, completely open is not an option, either. Societal systems derive theirexistence from a recognizable identity, and identity assumes or requires distinction from its environment. So one might say that systems and theirenvironments together “constitute” these boundaries.

Subsystems

One of the possibilities within system approaches is to define subsystemstogether forming the system as a whole. It seems logical in the context ofthe interactive governance of societal systems to take as a choice criterionthese interactions, defined as a subsystem (GI) between, on the one hand,the subsystem-to-be-governed (SG) and, on the other hand, the governingsubsystem (GS). [For sake of presentation we delete the addition “sub”; instead we speak of societal system and the three subsystems SG, GS andGI.] So we might extend the definition of governability by seeing it as theoverall capacity for governance of any societal entity or system consistingof SG, GS and GI. In practice this means that we consider the concept of governability as an integrated whole, while explaining that each of its components (SG, GS and GI) has a conceptual basis on its own. All threecomponents also have their own governability aspects. The SG we see as all societal processes and structural arrangements, activities that form andsurround what we call societal primary processes, such as creating a family,taking care of the sick, producing cars and catching fish. We do not intendto develop general theories for them, but our interest is in focusing on governability issues they may raise. The governing system (GS) consists of actors, entities and parties having varying potentials available for theirgovernance roles and tasks with regard to the SG. GI we see as the subsystem where all relations and interactions between SG and GS are concentrated. Although GI is to a large extent dependent on the way GSand SG are able and willing to interact, they may develop characteristics of their own, which in the interactive governance perspective will receive special attention.

For governance or governability purposes, all aspects of systems, includingscales, levels, boundaries and nestedness, can be “problematized” and consideredrelevant depending on the research purpose or practical application. It doesmake a difference from a governance point of view where a boundary canbe imagined as resilient or fluid, a boundary can be considered as soft, hardor permeable, and the influence of other system levels in the nested hierarchytaken into consideration or not.

4. Governability and the system-to-be-governed (SG)

The key consideration with any SG is to determine where and what makesit more or less governable. So we have to specify and narrow down what we see as relevant for conceptualizing SG in such terms, and the best wayto do this is to find one or more good theories or conceptual approaches.These have to be of an interdisciplinary nature as we want to look at SG inbroad terms as long as we don’t know where to find such potential govern-ability factors.

It speaks for itself that we take the central concepts we have developedso far as our analytical point of departure. This is in the first place the inter-action concept used at the actor and system level as a central one as well as the idea that modern societies are being characterized by the features diversity, complexity and dynamics, and preferably as a combination.

Using the interaction and the system concept as two main concepts forconceptualizing SG, we might take two analytical roads, comparable to the“old-time” distinction between micro and macro. One is building up from a micro-societal situation, using interactions around a societal activity or primary societal process such as teaching a class, selling a car, building abridge or catching a fish as a starting point. One can look out for the diver-sity, complexity and dynamics of such processes and broaden their scope till a level has been reached where asking governability questions starts tomake sense. The second one is to begin with a macro-system situation such as a whole fishery, an education or industrial system and break thisdown into smaller components, find out about their diversity, complexity anddynamics, again until a level has been reached where governability issues one wants to pursue arise.

Now, we can think ourselves lucky that there is a body of literature whereboth approaches have been practiced (“system-down” and “actor-up”), andin fact we used this ourselves in earlier research on the governance offisheries. This is the chain concept, where a commodity, in our case fish, isfollowed from the eco-system to the consumer, enabling analysis not onlyof ecological, social, economic, cultural, ethical and political aspects of fish chains by scholars from different disciplines but also of their diversity,complexity and dynamics (Kooiman et al. 1999; Kooiman et al. 2005). Weshall now discuss these two analytical strategies.

Commodity chains as an example of conceptualizing SG

Although any type of primary process could be chosen, we shall focus onwhat have come to be known as commodity chains. Other terms and analysesare global value chains, commodity networks, circuits and, in the French tradition, filières – all having special meanings, research communities, ideo-logical and political backgrounds (for an overview, see Bair 2005). Of thosewe take two as they show nicely a system-and-actor approach to an SG: theworld-system and the global (value) chain.

World-systems studies of commodity chains are the most macro, stress-ing their role in the structuring and reproduction of capitalist society andinternational division of labor. Commodity chains are the units of analysisof contraction and expansion processes over time (many of the studies arehistorical analyses), with notions of competition at their basis. In the lastfew centuries such cyclical patterns in commodity chains have taken placeseveral times, according to this theory. Globalization, according to it, is

not a recent phenomenon; there has been a world economic system since thesixteenth century. A recent study shows that commodity chain analysis “notonly can chart the growing disjuncture between global economic activity andglobal income distribution, but also can provide causal explanations for thisoutcome” (Kaplinski 2001: 117).

Global commodity (value) chain (GCC) studies emphasize the organiza-tional character of the economic integration taking place around commoditiesunder the influence of globalization. In an authoritative definition, a GCCconsists of

interorganizational networks clustered around one commodity orproduct, linking households, enterprises, and states to one anotherwithin the world-economy. These networks are situationally specific,socially constructed, and locally integrated, underscoring the socialembeddedness of economic organization.

(Gereffi and Korzeniewicz 1994: 2)

Value chain studies have contributed much to the knowledge of chain processes. An important distinction is between producer-driven, e.g. auto-mobiles, or buyer-driven, mainly agricultural products chains where theinfluence of supermarket giants and retailers is great. To this set of two athird was recently added, trader-driven chains, “where international tradingcompanies . . . are able to procure continuously specific volumes and qualitymixes” (Gibbon 2001: 351).

These two examples of chain studies show the possibility of studying aparticular SG in ways where their actor as well as their structural aspect canbe taken into account. In this case different chains show different types ofgovernability characteristics, where world-system approaches show broad patterns of governability issues in a longer time perspective, such as who gets what at a global scale, while value chain studies show such issues morerelated to specific chains, such as who controls and governs them. It speaksfor itself that other societal systems such as public health or internationalfinance can be analyzed in actor-up or structure-down terms, and show their particular governability issues depending on the scale or level one wantsto pursue.

5. Governability and the governing system (GS)

Governability from the point of view of the GS is the capacity to bring about,organize and carry out governance activities in the face of societal and natural diversity, complexity and dynamics. In the interactive governanceapproach, a conceptual framework has been developed for this purpose fromwhich we shall present two attributes here: orders and elements of governance

(Kooiman 2003; the attribute modes will discussed under the heading of GI,as in the context of governability they fit better there). This capacity appliesto all three of the major components – state, market and civil society – andto the hybrid forms among them as subsystems of GS.

Governability and subsystems of GS

States are still the most central and omnipresent societal governance sub-systems. They steer and control from local to international levels in diverseand complex ways. For all practical purposes, the concept of a homogeneoussocietal institution, denoted as “the state” and governed by uniform rules,has to be replaced by other models, allowing variety and differentiation aswell as certain degrees of independence and interdependence. Changes alsoshow the dynamics of the modern state, and “[w]hilst the state . . . may bein retreat in some respects, its activity may be increasing in others. And nowhere. . . has its key decision-making role been seriously undermined” (Müller andWright 1994: 1).

Markets, as governance institutions, also have their own diverse, complexand dynamic features. Williamson’s view of governance (1975), in which insti-tutional economics provide the institutional framework, consists broadly ofmarkets, hierarchies and mixed forms of them, through which transactionsare channeled. This amounts to getting away from general economic lawsexplaining market interactions, but showing some of their governance aspects.Next to this the field of economic sociology is the main contributor to insightsinto the market as a governance institution and its societal characteristics ina broad sense (e.g. Swedberg 2005).

The governance roles of civil society can be conceived of as a societal insti-tution for which the term “public domain” has been coined; interactionsbetween the media, interest groups, universities and social movements dis-cussing, criticizing and mobilizing more informal governance forces. Thereis a massive literature on the subject, e.g. about what belongs to it or not,what its societal contributions are, differences between north and south. The dynamics and balance between state and civil society are in constantflux such that “neither of the two can monopolize public life without provoking a reaction from the opposite realm to retain political space” (Biekart 1999: 36–7).

Hybrids between the three societal institutions receive much attention today. Where the state withdraws and leaves some of its servicing tasks tothe market, companies with private or mixed ownership are also common.For governability this hybridization of institutions and the way they are institutionalized on the borderline between state, market and civil society are important and are challenging issues to assess (Van Tulder and Van derZwart 2006).

Governability and attributes of SG

Elements of governance

In the interactive governance theory, three governing elements are distinguished:images, instruments and action. All three play a role: images as sets of ideaswhere those governing want to go, instruments giving these ideas substanceand making them operational, and action potential needed to support theseinstruments and let them do their work.

Images constitute the guiding lights for the how and why of governance.Images come in many types: visions, knowledge, facts, judgments, pre-suppositions, hypotheses, convictions, ends and goals. They not only relate tospecific issues but also contain assumptions about fundamental matters suchas the relationships between society and nature, the essence of humankind, andmajor societal trends. For governance and for governability, it is importantthat images help in understanding the diversity, complexity and dynamicsof the subjects and issues they are facing. A distinction between types of images/knowledge might be helpful: ideas, arguments and data. The more these threesupport each other, and the more other actors are able to assess the validityof governance images from their perspective and to add to or modify it, themore they are shared, the more governance images can contribute to govern-ability. The same type of reasoning can be developed for the instrumentaland action component of governance (see further Kooiman 2003, 2008).

The governance elements images, instruments and action are highly inter-related. It is the fit between the three elements of governance that can beseen as the major contribution of these elements to governability.

Orders of governance

The theoretical framework developed here also relates to orders of govern-ance. These can be imagined as three concentric circles nested as in the peelof an onion. The three orders are closely related and always – even whenthey are not made explicit – available.

First-order governance takes place wherever people and their organizationsinteract in order to solve societal problems and create new opportunities. Itprovides the means of solving the constant stream of problems which surfacein the system-to-be-governed – problems of supply, price, market, employ-ment, work satisfaction, etc. In diverse, complex and dynamic societies, first-order governance faces special challenges. It starts with the identification of problems, a process which takes place first of all in the minds of societalactors. The first step in the governance process is therefore the identificationand formulation of societal problems, whereby the latter are distinguishedfrom private problems by their scale and shared nature. Once problems, andproblem systems, have been identified, attention shifts to the solution space.

It is important throughout to retain in the analysis the diversity, complexity,dynamics and scale of situations, as only then will images remain close toreality.

Second-order governance focuses on the institutional arrangements withinwhich first-order governing takes place. Here the term “institution” denotesthe agreements, rules, rights, laws, norms, roles, procedures and organizationsthat are applied by first-order governors to make decisions. Institutions provide the framework within which first-order governance takes place, andconstitute the meeting ground for those being governed and those govern-ing. Second-order governance implies the reconsideration and adaptation of the parameters of first-order governance.

Meta- or third-order governance feeds, binds and evaluates the govern-ing exercise. Many principles govern activities. For example, the prin-ciples responsibility and sustainability are recognized almost universally. Inmeta-governance, governors and governed alike take each other’s measurein formulating norms by which to judge each other and the measuring process, too.

For governability, the three orders have special importance: together theyform the backbone of the capacity of a GS to govern in the three dimen-sions sketched. Not only each by itself but crucially are the three governingorders or differentiated tasks connected with them in a societal system: arethey complementary to one another, or are they at odds?

It will be quite clear that this discussion of the three subsystems of GSand the choice of a few attributes is only the very beginning of analyzingand assessing GS. The attributes presented are from earlier work in the fieldof governance, and can be worked out and added to at will. The potentialliterature on these subjects is almost unlimited, and for each special case reasonable choices can be made. Basic, however, is the idea that the govern-ance capacity of SG and its contribution to the governability of a specificsociety or societal field has to include all three subsystems and, as we proposedfor SG, has to include structural- as well as action-level variables, preferablyin their mutual relation.

6. Governability and governance interactions (GI)

Interactions between SG and GS are crucial for governance, and thus import-ant for studying and assessing governability. Those governed, through theirparticipation, try to exert influence on those governing, and governing entitiestry to influence those governed through their policies and managementefforts. Above, the centrality of the interaction concept for our governanceperspective has been put forward; here we continue this presentation of ourideas at this point.

In the reality of modern governance, an enormous variety of interactionscan be observed. From the GI perspective, they can be ordered in a few major

types: participatory, collaborative and policy or management interactions (seeKooiman 2003 for the conceptual basis of this distinction).

For governability, it is important to know how social-political entities – such as individuals, organizations, groups, movements or other forms ofcollective action – participate in governing interactions. Where does such par-ticipatory action come from? Who acts and who reacts? I see participatoryinteractions as directed from SG to GS. The character of the interaction isdetermined by the responsiveness of those governing, on the one hand, and whathas been called the “repertory” of resources and activities which the governedcommand, on the other (Barnes and Kaase 1979). This repertory is wide and varies: voting, letter-writing, and protesting in sit-ins and boycotts, participating in a movement or being a member of a focus or action group, and many more. Social movements are the classical example of this kind of spontaneous, loosely organized form of governance interaction. As theseinteractions are the least systematically organized, I call them interferences.

The importance of collaborative forms of governance interactions is growing.Why, for governance purposes, are groups, organizations and authorities willing to share their activities and aim to do things together instead of doing them alone? Often mutual interdependencies are mentioned as the main reason for such collaborative or co-operative interactions. Partnershipsbetween public and private entities are a popular form of such collabora-tion. But collaborative interactions between companies and NGOs can alsobe found in many fields (see, for a recent overview, Glasbergen et al. 2007).As these collaborative interactions are mainly of a horizontal nature, theyare indicated as interplays.

Policy and management interactions are the collective variables for all interventionist interactions by GS aimed at having an impact on SG. Publicauthorities at all levels have numerous interactions, dressed in policy terms,at their disposal to bring about politically preferred societal changes (Mayeret al. 2005). Management is seen as a way to organize these interactions accord-ing to criteria of efficiency and effectiveness. Stake-holder identification, forexample, has become a popular (interventionist) tool in this respect (Bryson2004: 32–3). As basically these interactions are formally organized, I call theminterventions.

Social-political cultural traditions and power relationships also find their expression in governance interactions. For example, it is often said that“Anglo-Saxon” social-political culture does not stimulate formal interactionsbetween governors and governed, in contrast to the “Continental” tradition,where those are enabled and often institutionalized. Such differences mayalso explain why co-governing interactions, such as co-management schemesin fisheries, are more common in some political cultures than in others (Wilson et al. 2003).

This points to the analytical distinction in governance interactions madefor interactions generally: an action or intentional level and a structural or

contextual level. This distinction, although the subject of heated social sciencedebates, such as in terms of agency-structure, is a useful one. Any concep-tualization of the constituent actors in a governance interaction necessarilyinvolves an idea of its structural component.

At the structural level, three modes of governance can be distinguished:self-, co- and hierarchical governance. These three modes roughly correspondto the three interaction modes at the action level of governance: interferenceswith self-governance, interplays with co-governance, and interventionist oneswith hierarchical governance. All societies demonstrate mixes of these threegovernance modes, and all three modes contribute in specific ways to therole governance systems and governance interactions play in governability.

In modern society, self-governance refers to situations in which actors takecare of themselves, outside the purview of government. Liberal political thinking typically highlights societal self-governing capacities, while socialist-oriented ones downplay them. It must be emphasized here that self-governanceis not necessarily a government-created capacity, but comes about of its ownaccord. In fact, if a capacity for self-governance is not sustained, societal governance is an impossible task. The collective-action school has made the most systematic analysis of self-governance with regard to the exploita-tion of common-pool natural resources, such as capture fisheries (Ostrom1990).

The essential element of co-governance is that societal parties join handswith a common purpose in mind, and stake their identities and autonomyon this process. Much attention has been devoted to co-governance and tothe opportunities that it offers. In capture fisheries, a form of co-governancecalled co-management has been particularly influential (Wilson et al. 2003).Governance theory contains numerous manifestations of co-governance,including communicative governance, public–private partnerships, networks,regimes and co-management (Kooiman 2003).

Hierarchical governance is the most classical mode, and is characteristicfor the interactions between a state and its citizens, but also for companieswith their personnel. It is a top-down style of intervention, expressing itselfin policies, rules and regulations. Steering and control are the key conceptshere. In recent years, perceptions of hierarchical governance have becomeredefined under the influence of market ideas and concepts like managingand client orientation (e.g. Ferlie et al. 2005).

Systems of GI

Combining the action and structural level of governance, interaction systemsof them come into sight. For analyzing governability, it is important to getan idea what the features of GI systems might be next to those of GS andSG. To understand what is going on in these GIs, again we might use insightinto diversity, complexity and dynamics. There is an important argument for

taking the diversity of values, goals and interests of those involved in inter-actions into account. This points to processes of ordering and re-orderingof aspects of diversity. The complexity of GIs can easily be understood ifone takes a closer look at all the groups and organizations involved in specificgovernance fields. And their dynamics can be seen as a composition of forceswhich sometimes turn into gradual developments, but more often result innonlinear patterns of change. Insights in societal dynamics have direct orindirect relevance for governance and an assessment of the role of the GI ingovernability. For example, in his study of interactions between state andsociety, De Vries (2005) distinguished four (macro) types of policy interactionsbetween government and society in the Netherlands by cross-tabulating(dis)parity of power and authority (vertical, horizontal) and perceived inter-ests (antagonistic, congruent). These four types were characteristic only for a certain period during the last fifty years, and after some time changedinto another system of policy interaction – a transition which expresses thedynamic nature of governance interactions.

7. Conclusion

In this essay I sketched the main contours of the governability concept basedupon the interactive perspective on governance. In other publications someof its aspects have been dealt with more extensively, and interdisciplinaryactivities evolved to work with and test it in the study and practice of fisheries,coastal zones and aquaculture (see papers in special issue of Journal of Trans-disciplinary Environmental Studies, 2008: Kooiman et al. 2008; Chuenpagdeeet al. 2008; Mahon 2008; Bavinck and Salagrama 2008; also Jentoft 2007;Mahon et al. 2009; Jentoft and Chuenpagdee 2009). There are still many con-ceptual questions not answered. However, a systematic beginning has beenmade to release it from its earlier scholarly isolation by an unfortunate pseudo-political use of the concept. So far it has shown its utility in actual researchin the field of aquatic resources, and there is no immediate reason why itcould not be applied in other societal fields as well.

References

Bair, J. (2005) “Global Capitalism and Commodity Chains: Looking Back, GoingForward”, Competition and Change, 9 (2): 153–80.

Barnes, S. H. and Kaase, M. (1979) Political Action, Beverly Hills, Calif.: Sage.Bavinck, M. and Salagrama, V. (2008) “Assessing the Governability of Capture Fisheries

in the Bay of Bengal: A Conceptual Exploration”, Journal of TransdisciplinaryEnvironmental Studies, 7 (1).

Berger, P. L. and Luckmann, T. (1966) The Social Construction of Reality: A Treatisein the Sociology of Knowledge, Garden City, NY: Anchor Books.

Biekart, C. H. (1999) The Politics of Civil Society Building, Utrecht: InternationalBooks.

Bryson, J. M. (2004) “What to Do when Stakeholders Matter: Stakeholder Identi-fication and Analysis Technique”, Public Management Review, 6: 21–54.

Chuenpagdee, R., Kooiman, J. and Pullin, R. (2008) “Assessing Governability inCapture Fisheries, Aquaculture and Coastal Zones”, Journal of TransdisciplinaryEnvironmental Studies, 7 (1).

Cilliers, P. (1998) Complexity and Postmodernism, London: Routledge.De Vries, M. S. (2005) “Generations of Interactive Policy-making in the Nether-

lands”, International Review of Administrative Sciences, 71: 577–91.Ferlie, E., Lynn, L. E., Jr, and Pollitt, C. (eds) The Oxford Handbook of Public Manage-

ment, Oxford: Oxford University Press.Gell-Mann, M. (1994) The Quark and the Jaguar, London: Little, Brown.Gereffi, G. and Korzeniewicz, M. (eds) (1994) Commodity Chains and Global

Capitalism, Westport, Conn.: Praeger.Gibbon, P. (2001) “Upgrading Primary Production: A Global Commodity Chain

Approach”, World Development, 29 (2): 345–63.Giddens, A. (1984) The Constitution of Society, Berkeley, Calif.: University of

California Press.Glasbergen, P., Bierman, F. and Mol, J. P. A. (eds) (2007) Partnerships, Governance

and Sustainable Development, Cheltenham: Edward Elgar.Jentoft, S. (2007) “Limits to Governability? Institutional Implications for Ocean

and Coastal Governance”, Marine Policy, 4: 360–70.Jentoft, S. and Chuenpagdee, R. (2009) “Fisheries and Coastal Governance as a Wicked

Problem”, Marine Policy, 4: 553–60.Jervis, R. (1997) System Effects: Complexity in Political and Social Life, Princeton,

NJ: Princeton University Press.Kaplinski, R. (2001) “Globalisation and Unequalisation: What Can Be Learned from

Value Chain Analysis”, Journal of Development Studies, 327 (2): 117–46.Kjær, A. M. (2004) Governance, Cambridge/Oxford/Boston, Mass.: Polity.Kooiman, J. (ed.) (1993) Modern Governance, London: Sage.Kooiman, J. (2003) Governing as Governance, London: Sage.Kooiman, J. (2008) “Exploring the Concept of Governability”, Journal of Compara-

tive Policy Analysis, 10 (2): 171–90.Kooiman, J., Bavinck, M., Chuenpagdee, R., Mahon, R. and Pullin, R. (2008) “Inter-

active Governance and Governability: An Introduction”, Journal of TransdisciplinaryEnvironmental Studies, 7 (1).

Kooiman, J., Bavinck, M., Jentoft, S. and Pullin, R. (eds) (2005) Fish for Life,Amsterdam: Amsterdam University Press.

Kooiman, J., Vliet, M. van and Jentoft, J. (eds) (1999) Creative Governance, Aldershot:Ashgate.

Mahon, R. (2008) “Assessing Governability of Fisheries Using the InteractiveGovernance Approach: Preliminary Examples from the Caribbean”, Journal ofTransdisciplinary Environmental Studies, 7 (1).

Mahon, R., Fanning, L. and McConney, P. (2009) “A Governance Perspective onthe Large Marine Ecosystem Approach”, Marine Policy, 2: 317–21.

Mayer, I., Edelenbos, J. and Monnikhof, R. (2005) “Interactive Policy Develop-ment: Undermining or Sustaining Democracy”, Public Administration, 83: 179–99.

Mayntz, R. (2004) “Mechanisms in the Analysis of Social-macro Phenomena”,Philosophy of the Social Sciences, 34 (2): 237–59.

Müller, W. C. and Wright, V. (1994) “Reshaping the State in Western Europe”, West European Politics, 17: 1–11.

Osborne, S. (2006) “The New Public Governance?”, Public Management Review, 8 (3): 381.

Pierre, J. (ed.) (2000) Debating Governance, Oxford: Oxford University Press.Schuppert, G. F. (ed.) (2005) Governance-Forschung, Baden-Baden: Nomos.Swedberg, R. (2005) Principles of Economic Sociology, Princeton, NJ: Princeton

University Press.Van Tulder, R. and Van der Zwart, A. (2006) International Business–Society

Management, London: Routledge.Williamson, O. E. (1975) Markets and Hierarchies, New York: Free Press.Wilson, C. D., Nielsen, J. R. and Degnbol, P. (eds) (2003) The Fisheries Co-

management Experience, Dordrecht: Kluwer.

DOES GOVERNANCE EXIST?

Owen Hughes

The question in the title has a fairly obvious answer. Of course govern-ance exists, so the question posed is moot from the very start. Consult anydictionary, and the word appears. That there is a noun “governance” isundoubted; precisely what it means is contested – so contested that the wordhas lost much of its utility. Much has been written about governance, par-ticularly since its relatively recent rediscovery in the political science/publicadministration literature. As Frederickson has noted, “because govern-ance is a power word, a dominant descriptor, and the current preference of academic tastemakers, there has been a rush to affix to it all of the other fashions of the day” (2005: 285). Governance in some usage becomessomewhat mysterious; meanings are assigned that have relevance only for a cognoscenti socialized and imbued into the mystery, with everyone elseexcluded. This is problematic.

Much of what follows is about the word “governance”, and its meaningsand usage. It is argued here that it is the standard dictionary meanings of“governance” that have resonance, utility and acceptability, and to attachspecial meanings to an ordinary word is unhelpful. Governance is about running organizations, about setting up structures to enable the organiza-tion to be run. Although words may change meaning over time, to ascribeobscure meanings to a word that has a commonplace meaning that is differ-ent is to invite irrelevance. Aside from some fashionable usage, “governance”is a good word; it is a simple word with simple meanings, and should beused accordingly.

Standard meanings of governance

It is beyond doubt that “governance” is used more than it was in the politicalscience and public management literatures. However, as Pierre and Petersargue, “the concept of governance is notoriously slippery; it is frequently usedamong both social scientists and practitioners without a definition which allagree on” (2000: 7). A lack of definitional clarity may not be that unusualin the social sciences – does anyone agree on a single definition for “justice”

or “democracy”? – but usage of such terms is understood as requiring some explanation or qualification when they are used. For some reason, “governance” has become a word for which, despite the innate slipperinessof the concept, there are attempts – often wildly different – at precision indefinition or ascription of uses. It is argued here that these attempts do notadd much to understanding; the major reason being that, alongside the moreabstruse recent usages, there are standard meanings that are clear, generallyused and well understood.

At base, there is, in English, the verb “govern”, which derives from theLatin gubernare, meaning “steer, direct, rule”; and this, in turn, derives fromthe Greek kubernan, meaning “steer”. The verb “govern” can then attract thewell-known and common suffix “-ance” in order to transform the verb to a noun, in exactly the same way that “perform” becomes “performance”.“Governance” is but one of many nouns deriving from “govern”; others include“government”, “governor” and “governability”. After this start, though, thereis little agreement about governance.

The New Shorter Oxford lists three related meanings for governance: thefirst meaning is “the action, manner, or fact of governing; government” andalso includes “controlling or regulating influence, control, mastery” and “thestate of being governed; good order”. The second meaning is “the function orpower of governing; authority to govern” and “a governing person or body”;and the third meaning is “conduct of life or business, behaviour”. These mean-ings are not inconsistent with each other, nor with other dictionaries. They havealso been reasonably settled for several centuries: the first meaning is fromMiddle English, the second from later Middle English to the late sixteenthcentury, and the third from late Middle English to the mid-seventeenth century.

The dictionary meanings are also quite consistent with ordinary usage, especially “controlling or regulating influence, control, mastery” and the “conduct of life or business”. Governance is about running organizations,about steering as in the original derivation, how to organize, and how to setprocedures for an organization to be run. In what follows, this meaning is referred to as the standard, dictionary sense. In ordinary parlance, it isquite normal and correct to refer to the governance of schools and golf clubs,corporations, universities and even entire societies.

Not only is governance most often used in this steering and running organizations sense, it is also used this way in academic circles. Kjaer, for instance, sees governance as “the setting of rules, the application of rules,and the enforcement of rules” (2004: 10). Donahue argues that account-ability underpins civilization and that “one broad model of accountability is governance – the rules and institutions for the authoritative organization of collective life” (2002: 1). Here, governance is taken in its broad meaning andis entirely consistent with the standard dictionary definition and usage.

Kooiman, too, is quite careful in his usage of terms. What he terms “social-political governance” is defined as “all those interactive arrangements in

which public as well as private actors participate aimed at solving societalproblems, or creating societal opportunities, and attending to the institutionswithin which these governing activities take place” (1999: 70). Kooiman also distinguishes between governance and governability; governance “canbe seen as the total effort of a system to govern itself; governability is theoutcome of this process – not an ‘end state’ but a stock-taking at a par-ticular moment in time of complex, diverse and especially dynamic processes”(1999: 87). Regarding governance as “interactive arrangements” and as “the total effort of a system to govern itself ” is entirely consistent with thestandard meaning, and with “social-political” setting the context. In a later work, Kooiman draws a distinction between governing and governance,as follows:

Governing can be considered as the totality of interactions, inwhich public as well as private actors participate, aimed at solvingsocietal problems or creating societal opportunities; attending to the institutions as contexts for these governing interactions; and establishing a normative foundation for all those activities.

Governance can be seen as the totality of theoretical conceptionson governing.

(2003: 4)

This does provide an additional layer of abstraction in that there are clearlymeanings of governance that do not require them to be about theoreticalconceptions. However, Kooiman’s work is argued to draw quite explicitlyfrom the “steering” meaning of governance through the Dutch besturen andis entirely consistent.

Government and governance

As mentioned earlier, there is one, perhaps rare, point of some agreementamong commentators and this is that “government” and “governance” nowhave different meanings. As Rhodes argues, “current use does not treat governance as a synonym for government” (1996: 652). Pierre and Peters,too, argue that government and governance may have the same derivation,but “they need not, and indeed, should not, be taken to mean the same thing”(2000: 29). However, some works do conflate the two.

An OECD paper defines governance as “the formal and informal arrange-ments that determine how public decisions are made and how public actionsare carried out, from the perspective of maintaining a country’s constitutionalvalues in the face of changing problems, actors and environments” (2005:16). This definition is only about public governance and is, therefore,already an unsatisfactory definition for governance as a whole. The paperthen argues:

The 30 member countries of the OECD share core governance ele-ments. These have emerged with the evolution of the modern stateand include: democracy and citizenship; representation; a constitu-tion; the rule of law; competitive party and electoral systems; a permanent civil service; separation of powers between the executive,the legislature and the judiciary; and secularity.

(OECD 2005: 15)

A footnote here says “drawn from Finer (1997)”. Now, it is the case that Finer does mention all these things, but quite clearly in the context of government – the focus of his magisterial work. Finer’s work is explicitlyabout the state, the authoritative forces of the police and the military.

In an earlier work, Finer found four meanings for “government”. The firstis “the activity or the process of governing, i.e. of exercising a measure ofcontrol over others”; the second is “a condition of ordered rule”; the thirdis “the people charged with governing”; and the final one is “the manner,method or system by which a particular society is governed” (Finer 1970:3–4). And, as he continues, “this is all fairly straightforward because eachof these meanings is well accepted and the only problem is to sort out thedifferent usages and keep them separate” (1970: 4).

There are two points that follow. First, “government” is indeed about control, about authority, about ordered rule, as well as about those peopleable to do the ordering and the ruling. This part is quite clear. Second, thelast meaning of Finer’s, “the manner, method or system by which a par-ticular society is governed”, would, in current usage, be “governance” ratherthan “government” (see Rhodes 1996).

A government is the formal apparatus of society and can impose its willon the society. Governments have force at their disposal; they can requirecompliance through laws, and the coercion implied by those laws can be carried out. In societies with some pretence at democracy, the use of forcemay be muted – the Army and the police are nowhere near large enough tomaintain the regime through force alone – and the legitimacy of governmentmaintained by some kind of popular sovereignty. But the key point aboutgovernment is that force is ultimately behind it; no other lawful authoritybut government can compel people to act in ways that it prescribes. Govern-ment and governance are not the same; that much can be agreed. There can,indeed, be governance without government.

Rhodes (1996) and Bevir and Rhodes (2003)

Usages of governance, consistent with standard meanings, exist alongside far more esoteric ones. The question that follows is: Do these add to under-standing about governance as opposed to that obtained from ordinary,standard usage? While there are many formulations, one of the most widely

cited formulations on governance is that of Rhodes (1996), who found whatwas termed “six separate uses of governance”. These are:

• as the minimal state;• as corporate governance;• as the new public management;• as “good governance”;• as a socio-cybernetic system; and• as self-organizing networks (Rhodes 1996: 653).

In a later formulation by the same writer as a co-author (Bevir and Rhodes2003), there are seven meanings; now – significantly – termed “definitions”rather than “uses” as in the earlier work. Most are the same as the Rhodes(1996) list, although the minimal state “use” is no longer present, and twonew definitions have been added: “governance as international independence”and “governance as the new political economy” (Bevir and Rhodes 2003:45–53). Some of these provide real insight, but others are neither meaningsnor definitions. Using the earlier, dictionary definition of governance as running things and organization to run things, many are simply not required,that is, they are ordinary uses of the ordinary meanings of governance.

To start with the original Rhodes (1996) list, governance “as the minimalstate” is argued by Rhodes to be about redefining public intervention andusing markets to deliver services. He quotes Stoker approvingly that “govern-ance is the acceptable face of spending cuts” (1996: 653). Rhodes argues that this kind of governance encapsulates the ideological preference for less government, but that this “says little else being an example of political rhetoric”(1996: 654). It is difficult to see any definition of governance in this point;indeed, it is dismissed by Rhodes, and it is hardly surprising that the notiondid not survive to the Bevir and Rhodes (2003) list.

Similarly, the “governance as good governance” definition is really an ordinary use of the term with a commonplace qualifying adjective. Bevir and Rhodes argue that good governance refers to international agencies such as the World Bank in their promoting of better governing, and that“good governance tries to marry NPM to the advocacy of liberal democracy”(2003: 47). Presumably this means that good governance is about inter-national agencies packaging up NPM with liberal democracy and selling it to developing nations under the guise of good governance. It is an odd claim given that there has been great caution in the application of NPM in such contexts and that building institutional capacity has been the main emphasis (Minogue et al. 1998). More apposite here is Kettl’s argu-ment that the management reform movement “builds on the notion that good governance – a sorting out of mission, role, capacity, and relationships– is a necessary (if insufficient) condition of economic prosperity and socialstability” (2005: 6).

However, it is argued there is no new definition of governance in the notionof good governance. Certainly, the World Bank and other international insti-tutions have become more interested in how countries, particularly countriesreceiving aid, organize and run themselves. They have become interested ingood governance, that is, that societies are well run. But, again, all that ishappening here is that the ordinary meaning of governance is being used.

Governance “as international independence” is similarly not a meaningand even less a definition. For Bevir and Rhodes, this involves the weaken-ing of the nation-state’s capacities for governance, through hollowing out of the state and multi-level governance (2003: 47). While there is extensivediscussion of governance issues by scholars of international relations, there is no need for a special usage or meaning to be attached to the use of governance in such a context. Rosenau, for instance, from this inter-national relations literature, argues that governance refers to “mechanismsfor steering social organizations toward their goals” (1997: 40), also “the process of governance is the process whereby an organization or society steers itself ” (1997: 146). Both points use the ordinary “steering” definitionof governance and apply it to the running of entire societies and nation-states. It is argued that governance “as international independence” is nota useful definition.

Rhodes and Bevir and Rhodes argue that the power available to central-government actors has declined and that one definition of governanceresulting is “as a socio-cybernetic system”. In this definition of governance,“there is no longer a single sovereign authority and blurred boundaries betweenpublic, private and voluntary sectors”, with examples cited including “self-regulation and co-regulation, public–private partnerships, co-operativemanagement, and joint entrepreneurial ventures” (2003: 48). There may be a valid point here, but it does not add up to a definition as such. It israther a recognition that governance needs to be looked at in its wider andoriginal sense of steering, rather than as a synonym for government. Also,it is an unhelpful term that Rhodes has invented for the purpose. Given that“cybernetics” derives from the same Greek and Latin as does “governance”(Rosenau 1997: 146), calling governance “a socio-cybernetic system” is anobvious tautology. It is also unhelpful to define governance in relation toanother term – cybernetics – that similarly evades agreed definition.

It is also difficult to see a distinction between this definition and that of governance “as the new political economy” which “re-examines both thegovernment of the economy and the boundaries between civil society, stateand the market economy as they become increasingly blurred” (Bevir andRhodes 2003: 48–9). With blurred boundaries in this definition and the socio-cybernetic one, surely there is no separate meaning, either. In both ofthese, the site for governance cannot be located with certainty. As Kennettargues, “with the change from government to governance the governing administration is now only one player amongst many others in the policy

arena” (2008: 4). There may be a real point to some of the ideas in “socio-cybernetic governance”, but it is not a definition of governance, nor is it necessary to have one. Even if “socio-cybernetic” can be accepted as beingmeaningful, it is no more than a qualifying adjective to the standard mean-ing of governance. Other actors may well exercise governance functions, but such activities can co-exist within the normal meaning of governance assetting up a set of rules to run organizations.

There are other parts of the Rhodes (1996) and the Bevir and Rhodes (2003) lists of definitions that do warrant further discussion: governance ascorporate governance; governance as the new public management (NPM);and governance as networks.

Corporate governance

One of the Rhodes and Bevir and Rhodes definitions of governance is “ascorporate governance” (Bevir and Rhodes 2003: 45–6). A minor point is thata word should not be defined in terms of itself; but, more broadly, it is arguedthat this is not a definition at all, rather an ordinary usage of running thingsand steering, in this case involving corporations.

There is a large and important literature on corporate governance. Thereare many academics writing and researching issues about the running of corporations, in particular their accountability structures, principal–agent issues and the like (Jensen 2000; Keasey et al. 2005b). Moreover, there has been more action, more serious theorizing in corporate governance inthe private sector than in public-sector governance in recent years. From principal–agent theory to transaction–cost theory to Enron and Sarbanes/Oxley, questions of corporate governance have dominated management in the private sector. Of course, there may be a tangential governmentinvolvement here in that the corporation’s legal environment as set out by government is a key consideration in issues of corporate governance. But it is not the only consideration and neither is it necessarily the most important. The more recent concerns about corporate governance havebeen mainly about designing internal and external structures of accountabilitythat lead to good managerial performance. The motivation is mainly to main-tain and enhance shareholder and investor confidence rather than to satisfyany direct requirements from government.

However, corporate governance does not require a separate definition of governance. Its usage in this context is simply “governance” being usedin its standard meaning. For instance, the Cadbury report in the UK in 1991 defined corporate governance as “the system by which companies aredirected and controlled” (Keasey et al. 2005a: 21). An Australian govern-ment report defines corporate governance as encompassing “the arrangementsby which the power of those in control of the strategy and direction of anentity is both delegated and limited to enhance prospects for the entity’s

long-term success, taking into account risk and the environment in which it is operating” (Australia 2003: 2). The paper continues by arguing that agood governance framework “should guide the actions of individuals by providing clarity of direction as to appropriate behavior and decision-making”(Australia 2003: 2). Governance is also seen as being about accountabilityin that “a robust governance framework should, through transparency andaccountability mechanisms, link power and responsibility to performance andreview” (Australia 2003: 2–3).

Not only are these points a good exposition of corporate governance; they are also entirely consistent with the standard meanings of governanceoutlined earlier. Applying governance to a corporate context does notrequire another definition at all; it is still about steering mechanisms but ina different context – the corporation rather than the society.

Governance as the new public management (NPM)

Another definition of governance in Bevir and Rhodes is “as the new publicmanagement”. The key example used by them is the distinction made in akey work between steering and rowing, as they argue:

Osborne and Gaebler distinguish between “policy decisions (steer-ing) and service delivery (rowing)”. They argue bureaucracy is abankrupt tool for rowing. In place of bureaucracy, they propose“entrepreneurial government” which will stress competition, markets,customers and measuring outcomes. This transformation of thepublic sector involves less government (or less rowing) but more governance (or more steering).

(Bevir and Rhodes 2003: 46)

Two questions follow from this. First, does governance as used by Osborneand Gaebler or NPM constitute a new definition of the word “governance”;and, second, has governance been a major concern of NPM, sufficient foran entire new definition to apply?

First, the distinction between steering and rowing was actually made bySavas (1987) and is duly attributed by Osborne and Gaebler (1992). Even ifthe latter’s use has had more currency, it is misleading to characterize themas simply being about steering and rowing. Also, at no point do Osborneand Gaebler even mention NPM. Their ten principles of “entrepreneurial government” are quoted by Rhodes, who then argues “clearly, NPM andentrepreneurial government share a concern with competition, markets, customers and outcomes” (1996: 46). There may be similarities, but it is simplistic to conflate the two.

Moreover, when Osborne and Gaebler do use “governance” it is done inquite a standard way. For instance, they argue (with original emphases):

Our fundamental problem is that we have the wrong kind of govern-ment. We do not need more government or less government, we needbetter government. To be more precise, we need better governance.Governance is the process by which we collectively solve our prob-lems and meet our society’s needs. Government is the instrumentwe use. The instrument is outdated, and the process of reinventionhas begun.

(1992: 23–4)

At another point, Osborne and Gaebler (1992) argue that privatization forideological reasons is selling snake oil:

Services can be contracted out or turned over to the private sector.But governance cannot. We can privatize discrete steering functions,but not the overall process of governance. If we did, we would haveno mechanism by which to make collective decisions, no way to setthe rules of the marketplace, no means to enforce rules of behavior.

(1992: 45)

There is no new use of the word “governance” in Osborne and Gaebler, no novelty, certainly no definition, only the use of the word in its standarddictionary sense. Both Rhodes and Bevir and Rhodes argue that NPM isrelevant to the discussion “because steering is central to the analysis of publicmanagement and a synonym for governance” (Rhodes 1996: 655; Bevir andRhodes 2003: 46). This could be argued to prove the point that governanceitself is about steering; Osborne and Gaebler are doing nothing more thanusing the term quite correctly according to its standard dictionary definition.

The second point to look at is the extent to which NPM is or was aboutgovernance in any case, sufficient for NPM then to be a definition of govern-ance. It is argued here that, while there may be some system design aspectsof NPM involving governance, these have not been anywhere near as sub-stantial as the concerns in NPM about managing within government.

The most profound change resulting from NPM-type reforms has been therecognition that public managers are themselves responsible for achievingresults and must take personal responsibility for doing so (Hughes 2003).From this change alone comes the need to decide if results have beenachieved – performance measurement in its various permutations – as wellas changes in accountability and much else. These are largely internal. AsKickert argues:

Management in the public sector has to deal with more value patterns than business-like effectiveness and efficiency criteria, suchas legality and legitimacy, social justice, and equal rights. So, we needto broaden the concept of public “management” to “governance”.

This concept is not restricted to internal management and organiza-tion but explicitly encompasses external relations between public organizations and their socio-political environment.

(Kickert 1997: 197)

NPM was always more about the inside of the organization as well as enhancing the personal responsibility of managers. This was to occur withthe governance structures largely unchanged; the governance part of NPMwas minor and tangential rather than central. Therefore, governance “as NPM” is neither necessary nor helpful in a terminological sense.

To complicate the issue further, there are major problems in deciding exactly what is involved in NPM. It is often naïvely and incorrectly seen as a program. Not only is there no agreed set of points constituting NPM;neither is there anyone who puts the so-called program forward. And, as Hoodand Peters argue, “Like most divinities, NPM turned out to be somewhatmystical in essence, as no two authors of that era listed exactly the same features in enumerating its traits” (2004: 268). Fountain describes NPM as “a loose collection of policy and management initiatives” (2001: 19);Christensen, Lie and Lægreid similarly argue that NPM “is actually a rather loose concept encompassing several different administrative doctrines”as well as being “rather contradictory” (2007), and Van Thiel, Pollitt andHomburg argue:

NPM is like a chameleon: it constantly changes its appearance to blend in with the local context. . . . Such adaptability is possible,because NPM is not a coherent set of ideas and tools. The ideas mightbe the same, but the underlying story differs all the time.

(2007: 197)

All of these writers admit, at least implicitly, that they cannot define NPM.What we have is a theory without a theorist, a program without an advo-cate and an agenda without anyone putting it forward. In so far as there isany agreement as to what is involved in NPM, the most salient points arewithin the organization and not in system-wide changes to governance.

Two other uses of “governance” in the context of public management needto be mentioned. These are put forward as replacements for NPM – one fromDunleavy et al. (2006) and the other from Osborne (2006).

Dunleavy et al. (2006) argue that NPM is dead and has been overtakenby what they term “digitally enhanced governance” (DEG). This is char-acterized by “reintegration”, “holism” and “digitalization”. Reintegrationmeans rolling back the disaggregation of NPM to include such points as joined-up governance and “re-governmentalization”. Holism is to include client-based or needs-based reorganization, one-stop provisions, ask-once pro-cesses, interactive and “ask-once” information-seeking, data warehousing,

pre-emptive needs analysis, end-to-end service re-engineering and agile government processes. Digitalization elements include electronic servicedelivery and e-government, new forms of automated processes, radical dis-intermediation, active channel-streaming, customer segmentation, mandatedchannel reductions in order to move toward co-production of services, quasi-voluntary compliance, do-it-yourself forms and tax-paying and open-bookgovernment. But, even if there is no real explanation as to how digitallyenhanced governance is different from government, there is no inkling thatDunleavy et al. are using governance in an unusual way that requires anynew definition.

Another paper, by Osborne (2006), argues for something new called New Public Governance (NPG). Osborne argues that the time of NPM is“a relatively brief and transitory one between the statist and bureaucratictradition” of public administration and “the embryonic plural and pluralisttradition” of what is termed New Public Governance. NPG posits “both aplural state, where multiple inter-dependent actors contribute to the deliveryof public services and a pluralist state, where multiple processes inform thepolicy making system” and as a “consequence of these two forms of plural-ity, its focus is very much upon inter-organizational relationships and thegovernance of processes, and it stresses service effectiveness and outcomes”(Osborne 2006).

Even if NPG is thus far diffuse and ill-defined, it may well provide a wayforward, particularly with its clear complementarity with Kooiman’s “socialpolitical governance”. However, NPG does need to avoid a programmaticapproach. In the same way that failure to establish what the NPM programwas doomed it as a movement, a similar fate may well occur for NPG. But,for present purposes, Osborne, too, does not need a new definition of govern-ance that relies on meanings beyond the standard dictionary definition.

Governance, networks and markets

One particularly contentious definition of governance is that it is all aboutnetworks. Rhodes argues that “governance has too many meanings to be use-ful, but that the concept can be rescued by stipulating one meaning. . . . So,governance refers to self-organizing, inter-organizational networks” (1996:660). Two points follow from this. First, the governance-as-networks defini-tion must be all-encompassing and sufficiently robust to capture all behaviorsinvolving what was previously known as governance. Second, if there is tobe only one meaning, then there is a presumption that all others need to bediscarded, including the standard dictionary definitions discussed earlier.

On the first point, networks are argued by Rhodes to be a new form oforganization, a third way of organizing, and that “to markets and hierarchies,we can now add networks” (1996: 653). Governance is now to be defined“as networks”:

From this perspective, governance consists of self-organizing, inter-organizational networks. These networks are characterized, first, by interdependence between organizations. Changes in the role of the state mean the boundaries between the public, private andvoluntary sectors are shifting and opaque. Second, there are con-tinuing interactions between network members, caused by the needto exchange resources and to negotiate purposes. Third, these interactions resemble a game with actors’ behavior rooted in trustand regulated by rules that are negotiated and agreed by networkparticipants. Finally, the networks have a significant degree ofautonomy from the state. Networks are not accountable to the state;they are self-organizing.

Networks are argued to be a means for co-ordinating and allocating resources– a governing structure – in the same way as markets and bureaucracies. Thisis contrary to the usual distinction made between two forces: bureaucracyand markets (Ostrom 1974).

Adding a third player is an important claim, and part of an ongoing dis-cussion of the relative importance and interactions of governance, networksand markets. Where the network model is of some utility is in describing thepolitical machinations, the interest groups, the individuals and the inter-actions that lead to either an edict or a contract. Depending on point of view,the locus of government could be in the politicking before that point or theexercise of power itself.

Mintzberg argues that in the network model “government is viewed as oneintertwined system, a complex network of temporary relationships fashionedto work out problems as they arise, linked by informal channels of com-munication” (2000: 76). Kamarck also sees three actors, with the market as the third:

In entrepreneurial government the public’s work is done by peoplewho work for the government; in networked government the public’s work is paid for by the government even though it is per-formed by people who do not work for the government. In the thirdemerging model of government – market government – the work of government involves no public employees and no public money.In market government, the government uses its power to create amarket that fulfills a public purpose.

(Kamarck 2002: 249–50)

Bevir and Rhodes argue that “although the state does not occupy a priv-ileged, sovereign position, it can directly and imperfectly steer networks” andthe “key problem confronting government is, therefore, its reduced ability

to steer” (2003: 53). The reduced ability to steer is somewhat overstated. Itis the case that “government action depends increasingly on nongovernmentalpartners, from nongovernmental organizations that deliver public servicesto private contractors who supply important goods” (Kettl 2005: 6). But thisdoes not mean any necessary decline in the power of government at all, ratheran increased realization that the formal, rational bureaucratic model is nolonger appropriate and that it actually suits governments to involve a widerange of actors in what they do. This involvement may be aimed at makinggovernment more efficient and effective rather than ceding any real powerto the outside at all.

Take, for example, a police force or a tax office as a governmental organ-ization that has substantial, real power to force citizens to comply with rulings. Once, during the apogee of the traditional bureaucratic model, suchorganizations operated with little regard for the outside. The bureaucraticmodel assumes that all information is held inside and the traditionalbureaucrat makes a decision in an entirely rational way. But agencies thatact in a high-handed manner can lose public support even as they actentirely within their legal powers. And an agency that loses support may lose its “authorizing environment” (Moore 1995) and have its status and standing decline in the public mind.

Those in charge of the tax office now realize that the views of clients, account-ants and other players are quite valid inputs and, without changing its powersone iota, actively solicit opinions from outside players. There may be a net-work operating here; but the point is that, far from the network being in aposition of governing, it is rather that the exercise of power by governmentcan be better-directed with active outside involvement. As Kettl argues:

In order to embrace the large and complex networks responsible forservice delivery, many reformers now speak of governance insteadof government. As these networks have become more important, government officials have increasingly reached out to sweep theminto the reform movement as well. Improving government servicesrequires more than managing government agencies.

(Kettl 2005: 6)

On a contentious issue such as accounting standards, the outside pro-fessional body may have substantial input into legislation, but rather thanthe government conceding power it is rather exercising its unchanged powermore judiciously. It is no concession of power to a network to involve it in decision-making. Other organizations are assisting government in doingwhat it wants done and, “despite the view of some who persist in seeing networks as a weakening of the state, networked government can also be looked at as a different way of implementing the goals of the state”(Kamarck 2002: 246).

Pollitt is not overly impressed by the network approach, finding it “unsatisfying, indeed, at times, frustrating” (2003: 65). He does not like the“ahistorical assumption” that networks are new; nor the assertion that thereare now more networks or that these are somehow more democratic; northe methodological and theoretical weaknesses (2003: 65–6). As he argues, thebiggest disappointment for the public manager is “the paucity of interestingand tested propositions for action coming from network theory” (2003: 66).This is a particularly important point. If networks are to be a form of govern-ance, more needs to be demonstrated than that there is a lot of interactiongoing on between interests and governments. In sum, Pollitt argues:

This is not to say that network theory has nothing to offer. Rather,I see it as just somewhat overblown. At its worst it offers a romanticvision of a whole new way of governing, pivoted on a disparate bunchof case studies, usually drawn from a limited number of sectors. Atits best, though, it reminds us of the (longstanding) importance ofinformal relationships between organizations and groups, and howthese organizational dynamics can set the context for more formaland specific processes of decision-making.

(2003: 67)

Mintzberg, too, argues that reliance on the network model “can be over-done” and refers to France where “both public and private sectors have long been dominated by a powerful and interconnected élite that moves around with a freedom and influence that is proving increasingly stifling tothe nation” (2000: 78–9). The same general point could be made for otherEuropean countries.

It is quite obvious that there is a lot of political bargaining at all levels ingovernments and societies. Networks exist and do have influence. But thereis some distance from seeing the bargaining in action, influencing public policy as a result, to saying that networks are running the government. Noris there any evidence that governance means self-organizing networks andthat this now needs to be the sole meaning of the word.

The second point is whether or not the case has been made for discard-ing all other definitions of governance than governance as networks. Evenwith due allowances for hyperbole, the network model is not sufficiently robustto have moved the English language away from the ordinary dictionary meanings of governance discussed earlier. To regard networks as governanceor governance as networks is to see the two terms as entirely synonymous.Rhodes argues that only one meaning of governance is to be accepted, thatis governance as self-organizing, interorganizational networks. This meansthat other usages, the several hundred years of reasonably settled meaning,the ordinary usage in terms of a school or a tennis club, the ordinary usagein a specialized context, such as corporate governance, or good governance,

all need to be discarded. There is some utility to the network approach eitheras an organizing principle or as one approach to governing, but to say thatthere is only one meaning of governance and that is as networks is, indeed,overblown.

A search for meanings

Social science is replete with words that seemingly defy clear meaning. Inhis 1970 book on comparative government, Finer argued that the study ofcomparative government “still wallows in semantic confusions”:

Some of the most commonly used words are homonyms – one iden-tical word with several meanings. Others are stipulatory: whetherone knows it or not, the writer is using the term in a special andoften highly personal way. Some terms are both.

(Finer 1970: 3)

Some current uses or definitions of “governance” are clearly stipulatory, someare highly specialized. Also, in the same era as Finer, Mills argued againstthose he termed “grand theorists” who “never get down from the higher generalities to problems in their historical and structural contexts” leadingto “unreality so noticeable in their pages” and

One resulting characteristic is a seemingly arbitrary and certainly endless elaboration of distinctions which neither enlarge our under-standing nor make our experience more sensible. This in turn isrevealed as a partially organized abdication of the effort to describeand explain human conduct and society plainly.

(Mills, 1970, p. 42)

Whatever else their contribution might be, some theorists of governance have quite certainly not obeyed the imprecation to plain explanation. Forinstance, Bevir and Rhodes state “we use an anti-foundational epistemologyand an interpretive approach to understand changes in British governance”(2003: 1). This is hardly about explaining human conduct and societyplainly. Further into their book, the abstraction goes further. As they argue:

Our decentred approach to governance casts a new light . . . becauseit treats bureaucracy, markets and networks as meaningful practicescreated and constantly recreated through particular, contingentactions. In many ways, therefore, networks are an integral part of our social and political life, they are characteristic of bureau-cracy and markets as well as governance. . . . Once we stop reifyinginstitutions, bureaucracies and markets, we find many of the

characteristics allegedly specific to networks are widespread aspectsof political structures. Our decentred approach encourages a shiftof focus from networks, now recognized as an integral part ofsocial life, to the beliefs held by actors and to the stories told bypolitical scientists. Governance is new because it signifies an import-ant change in these beliefs and stories.

The words are recognizably English but are stipulatory and lead to lots of subsequent questions. What does a decentered approach mean here, andhow does it cast a new light – and on what? What are “meaningful practicescreated and constantly recreated through particular contingent actions”?Exactly who “reifies institutions, bureaucracies and markets”, and to whatpurpose? What are the “beliefs held by actors”, and what kinds of “storiestold by political scientists” are to be read? More importantly, what does anyof this say about governance in Britain or anywhere else? It is hard to see whatthe student of governance will get from this, particularly public managerslooking for assistance to solve real problems (Sparrow 2000). Such use of lan-guage is more likely to widen the gulf between public management theory andpractice, and to separate political science from public management further.

Conclusion

“Governance” is a good word, one with many uses, but it is a word in need of rescue. It risks being so burdened with meaning that its ordinarystandard dictionary definition – entirely appropriate for many current uses– becomes lost. Governance is about running organizations, public and private; it is about steering; it is about solving societal problems. Govern-ance cannot be confined to the public sector; indeed, its greatest current usage is in terms of corporate governance rather than of public governance.Some definitions of governance pay insufficient attention – often none – toits usage and meaning in other contexts. Again, governance needs to be ableto include the rules for a tennis club or how the board of a company or aschool is to operate.

The definitions of governance as networks, as new public management, as socio-cybernetic systems, as the new political economy, to name but a few, do not add to understanding of governance as a word, although theymight have other utility. Governance should be used, governance should bediscussed, but let the purpose be clear and the explanation considered.

References

Australia (2003) Review of Corporate Governance of Statutory Authorities and OfficeHolders, Canberra: Commonwealth of Australia.

Bevir, Mark and Rhodes, R. A. W. (2003) Interpreting British Governance, London:Routledge.

Christensen, T., Lie, A. and Lægreid, P. (2007) “Still Fragmented Government orReassertion of the Centre?”, in T. Christensen and P. Lægreid (eds) TranscendingNew Public Management: The Transformation of Public Sector Reforms, Aldershot:Ashgate.

Donahue, John D. (2002) “Market-based Governance and the Architecture ofAccountability”, in John D. Donahue and Joseph S. Nye, Jr (eds) Market-basedGovernance: Supply Side, Demand Side, Upside and Downside, Washington, DC:The Brookings Institution.

Dunleavy, Patrick, Margetts, Helen, Bastow, Simon and Tinkler, Jane (2006) “NewPublic Management Is Dead – Long Live Digital-era Governance”, Journal of PublicAdministration Research and Theory, 16 (3): 467–94.

Finer, S. E. (1970) Comparative Government, Harmondsworth: Penguin.Finer, S. E. (1997) The History of Government from the Earliest Times, Oxford: Oxford

University Press.Fountain, Jane (2001) Building the Virtual State: Information Technology and

Institutional Change, Washington, DC: The Brookings Institution.Frederickson, H. George (2005) “Whatever Happened to Public Administration?

Governance, Governance Everywhere”, in Ewan Ferlie, Laurence E. Lynn, Jr andChristopher Pollitt (eds) The Oxford Handbook of Public Management, Oxford:Oxford University Press.

Hood, Christopher and Peters, Guy (2004) “The Middle Aging of New PublicManagement: Into the Age of Paradox?”, Journal of Public Administration Researchand Theory, 14 (3): 267–82.

Hughes, Owen E. (2003) Public Management and Administration, 3rd edn,Basingstoke: Palgrave Macmillan.

Jensen, Michael C. (2000) A Theory of the Firm: Governance, Residual Claims, andOrganizational Forms, Cambridge, Mass.: Harvard University Press.

Kamarck, Elaine Ciulla (2002) “The End of Government as We Know It”, in JohnD. Donahue and Joseph S. Nye, Jr (eds) Market-based Governance: Supply Side,Demand Side, Upside and Downside, Washington, DC: The Brookings Institution.

Keasey, Kevin, Short, Helen and Wright, Mike (2005a) “The Development of Cor-porate Governance Codes in the UK”, in Kevin Keasey, Steve Thompson and MikeWright (eds) Corporate Governance: Accountability, Enterprise and InternationalComparisons, Hoboken, NJ: John Wiley.

Keasey, Kevin, Thompson, Steve and Wright, Mike (eds) (2005b) Corporate Govern-ance: Accountability, Enterprise and International Comparisons, Hoboken, NJ:John Wiley.

Kennett, Patricia (2008) “Introduction”, in Patricia Kennett (ed.) Governance,Globalization and Public Policy, Cheltenham: Edward Elgar.

Kettl, Donald F. (2005) The Global Public Management Revolution, 2nd edn,Washington, DC: The Brookings Institution.

Kickert, W. (1997) “Public Governance in the Netherlands: An Alternative toAnglo-American ‘managerialism’ ”, Public Administration, 75: 731–52.

Kjaer, Anne Mette (2004) Governance, Cambridge: Polity.Kooiman, J. (1999) “Social–political Governance”, Public Management, 1 (1): 68–92.Kooiman, J. (2003) Governing as Governance, London: Sage.

Mills, C. Wright (1970) The Sociological Imagination, Harmondsworth: Penguin.Minogue, Martin, Polidano, Charles and Hulme, David (1998) “Introduction: The

Analysis of Public Management and Governance”, in Martin Minogue, CharlesPolidano and David Hulme (eds) Beyond the New Public Management: ChangingIdeas and Practices in Governance, Cheltenham: Edward Elgar.

Mintzberg, Henry (2000) “Managing Normatively”, in Henry Mintzberg and JacquesBourgault, Managing Publicly, Toronto: Institute of Public Administration Canada.

Moore, Mark (1995) Creating Public Value: Strategic Management in Government,Cambridge, Mass.: Harvard University Press.

Organization for Economic Cooperation and Development (2005) ModernisingGovernance, Paris: OECD.

Osborne, David and Gaebler, Ted (1992) Reinventing Government: How the Entre-preneurial Spirit Is Transforming the Public Sector, Reading, Mass.: Addison-Wesley.

Osborne, Stephen (2006) “Editorial: The New Public Governance?”, Public Manage-ment Review, 8 (3): 377–87.

Ostrom, Vincent (1974) The Intellectual Crisis in American Public Administration, revisededn, Tuscaloosa, Ala: University of Alabama Press.

Pierre, Jon and Peters, Guy (2000) Governance, Politics and the State, Basingstoke:Palgrave Macmillan.

Pollitt, Christopher (2003) The Essential Public Manager, Maidenhead: Open UniversityPress.

Rhodes, R. A. W. (1996) “The New Governance: Governing without Government”,Political Studies, 44 (4): 652–67.

Rosenau, James N. (1997) Along the Domestic–Foreign Frontier: Exploring Govern-ance in a Turbulent World, Cambridge: Cambridge University Press.

Savas, E. S. (1987) Privatization: The Key to Better Government, Chatham, NJ: ChathamHouse.

Sparrow, Malcolm (2000) The Regulatory Craft: Controlling Risks, Solving Problemsand Managing Compliance, Washington, DC: The Brookings Institution.

Van Thiel, S., Pollitt, C. and Homburg, V. (2007) “Conclusions”, in C. Pollitt, S.Van Thiel and V. Homburg (eds) New Public Management in Europe, Basingstoke:Palgrave Macmillan.

WHAT ENDURES? PUBLICGOVERNANCE AND THE

CYCLE OF REFORM

Laurence E. Lynn, Jr

In the field of public administration and management, something is alwaysnew. In America, proclaiming “the new” has long been popular among academics. John Gaus announced the “new administration” in the early 1920s;Leonard White celebrated the “new management” and John Pfiffner the “newpublic administration” in the 1930s. Participants in a historic conference atMinnowbrook unveiled a new “new public administration” in the 1970s. Andthe habit has spread to Europe. A “new public management” was famouslyidentified by Christopher Hood in the early 1990s and a “new city manage-ment” was born in the Netherlands.

As New Public Management (NPM) lost amplitude in the academic issue-attention cycle (Downs 1972) – pronounced “dead” by Patrick Dunleavy andcolleagues not two decades after its birth (Dunleavy et al. 2006) – increasingprominence has been given to the “next new thing”: the “new (public) governance”. As R. A. W. Rhodes has observed, the coming and going ofintellectual fashions is “often fun, sometimes instructive, rarely long-lived”(Rhodes 2000: 54). As such, new fashions might be little more than mani-festations of the sociology of academia. More charitably, they might be viewedas heuristics that stimulate critical thought and investigation, and innovationsin practice, although often at the cost of distorting history for effect.

Worth taking more seriously, however, is the implicit assumption that insti-tutions of governance are relatively malleable.1 “Each generation redesignsgovernment,” claimed a group of experts associated with America’s NationalAcademy of Public Administration (DeWitt et al. 1994: 175). This assertioncollides with a widely held contrary view that institutional evolution is generally path-dependent and incremental. How might we reconcile seeminglyconflicting claims of malleability and stability?

In taking up this issue, the meaning of terms is important. Define “insti-tutions” as “stable sets of commonly recognized formal and informal rules

that coordinate or constrain” behavior (Weimer 1995: 2 – 3); a “paradigm”as habits of thought or mind shared by policy-makers concerning how to resolve problems of governance (Margolis 1993); and “path dependence”as a process of narrowing the range of political and economic choices in a way that links decision-making through time (North 1990). Using thesedefinitions, the proposition that a new public governance is replacing what-ever preceded it can be restated as follows: The habitual reactions of policy-makers – legislators, elected executives, judicial officers, and administrators– to problems of governance are shifting irreversibly as a result of changesin the contexts of governing, and the result is or will be new, at least temporarily stable institutions. The conceptual questions raised by thisproposition are: What endures and what does not in national institutions of governance? and How and why do the answers matter? The relatedempirical question is: What kinds of evidence are necessary to sustain claimsthat “the new” is upon us?

This chapter addresses these questions and offers some speculative con-clusions. The following section considers the stories of two historical issue-attention cycles (Downs 1972) involving claims concerning the advent ofsomething new in state–society relations: American claims concerning the emergence of a new (public) administration at every level of government inthe 1920s and 1930s, and the claims first heard in the early 1990s concern-ing the emergence of a new (public) governance – a story that is ongoing.There follows a comparison of the arguments sustaining the two stories, raising the question of whether the second story might end as the first did:in the emergence of a genuinely new paradigm of administrative thought and action. The paper concludes with reflections on what can be said aboutwhat endures and what does not as fashions in public administration andmanagement reform succeed one another.

“The new”: old and new

An American penchant to proclaim or advocate “the new” had its first important expression following decades of Progressive-era reforms inAmerican institutions: the beginnings of a professional civil service; economicregulation; an income tax; the executive budget; new forms of direct demo-cracy; the direct election of US senators; and women’s right to vote. Fifteenyears later, the term “new” had been dropped, and “public administration”had become an accepted paradigm of thought and action.

Beginning in the twentieth century’s final decade, many on both sides ofthe Atlantic began claiming that a “new (public) governance” is replacingpublic administration as the profession’s foundational paradigm. For atime, this term competed for attention with the seemingly more resonant and popular NPM. Within a decade, however, the new (public) governancein various and diverse guises was ascendant.

Though this more recent discourse concerning “the new” has distinctivenational versions and has yet to run its course, comparing these two epi-sodes of the field’s intellectual and institutional history can be instructive.Unlike the reform cycle associated with NPM, the new (public) governancehas a reasonable but uncertain prospect of ending like the earlier Americanstory: in a new paradigm of thought and action.

The “New (Public) Administration”

In the latter decades of the nineteenth century, a modern administrative state began to replace America’s largely pre-bureaucratic, agrarian, populistinstitutions. Beginning with civil service reform in the 1880s, the pace of changeaccelerated, and government at all levels grew in scope, size and influence.The landmark Budget and Accounting Act of 1921 institutionalized modernexecutive government in America.

New problems

Leonard White summarized the context in which the American adminis-trative state emerged:

After the echoes of the Civil War had died away a new society was perceived to be emerging from the rural–small town economyof the first eighty years of our national existence. Invention was working its transformation through the industrial revolution and leading to an urban civilization based on the machine from the factory. This new civilization attracted millions of persons from other countries who came with different backgrounds of governmentand different standards of life. The new regime thus created new problems of public order, public health, housing, protection of safety standards, recreation, care of the unfortunate and the delinquent,regulation of utilities, and regulation of conditions of employment,and necessitated the development of new programs of governmentservice. . . .

(White 1933: 3)

Those who observed and participated in the reform process believed them-selves to be responding to unprecedented and dynamic social, economic, technological and political change that rendered nineteenth-century insti-tutions ramshackle and obsolete, not only in America but throughout theindustrializing world (Bertelli and Lynn 2006). Frederick Cleveland wrotein 1913: “A wave of organized democracy is sweeping the world, based ona broader intelligence and a more enlightened view of civic responsibility than has ever been obtained” (1913: 438). In the preface to his seminal 1926

PUBLIC GOVERNANCE AND THE CYCLE OF REFORM

textbook, An Introduction to the Study of Administration, White referred toadministration as “this enormous terrain, studded with governmental prob-lems of first magnitude” (White 1955 [1926]: xvi). During the administrationof President Franklin D. Roosevelt, White and his co-authors Marshall Dimockand John Gaus wrote:

The student of public administration is now confronted with a literally overwhelming torrent of new administrative agencies andnew problems to be assimilated and appraised. . . . [That] [s]o manycurrents of experience from industry, from psychological research,from the study of comparative government . . . to name but a few– are adding their contribution to the main stream require[s] bothimprovements in administrative technique and equally more accur-ate ideas concerning the nature of administration.

(Gaus, White and Dimock 1936: vii–viii)

A solution: public administration

In “The New Problem of Administration”, published in the early 1920s, JohnGaus argued that, faced with issues that were “technical, detailed, involved,[and] requiring special knowledge and expertise” (Gaus 1923–4: 220), politicalparties and legislatures tended “to leave difficult problems to administrativeagencies for solution” (Gaus 1923–4: 218). To the perfectly serviceable term “administration”, Gaus added the term “new”, defining “the newadministration” as a flexible instrument endowed with “a wide share of policy formulation . . . a large measure of discretion . . . [and] wide exemption fromjudicial review”, a transformation analogous to the creation of the “modernnational state” by European monarchs (p. 220). Gaus noted that the “new administration” had emerged “so suddenly that many of us are still unawareof its implications or problems or challenges” (p. 230).2

The cumulative result of these various transformations at local, state andfederal levels of American government was dramatic, wrote Harold Laski inthe new British journal Public Administration in the same year. “A state builtupon laissez-faire has been transformed into a positive state. Vast areas ofsocial life are now definitely within the ambit of legislation; and a correspondingincrease in the power of the executive has been the inevitable result” (1923:92). Pre-bureaucratic nineteenth-century patronage administration, frag-mented and decentralized, had been replaced by a modern, managerial andprofessional bureaucratic state with significant delegated authority to makepolicy, to regulate private activity, and to promote public welfare.

In 1933, White defined what he termed “the new management”, which, he said, emerged from 1900 to 1930, as “a contemporary philosophy of administration” favoring consolidation of administrative power at all levelsof government. The changes in administrative institutions during that period,

claimed White, were “primarily dedicated to greater efficiency, towardimproved methods, better and more extensive services and the eliminationof waste and irresponsibility” (White 1933: 4). Joining White in proclaim-ing “the new” was John Pfiffner. In his 1935 textbook, he wrote: “the newpublic administration is essentially democratic in spirit and practice. . . .[It] combines power with responsibility” to political officers who are underpopular control (1935: 19). Administration is “society working through govern-ment to solve the collective problems of a technological age” (p. 20).

By the mid-1930s, public administration had achieved a level of maturitythat no longer required the qualifier “new”; indeed, it was often referred toas “the” public administration. The field’s leading scholars and practitionersexpressed what had become “habits of mind”, a professional reasoning process that was concerned with a new agenda concerning

the interrelationships among the values of democracy; the dangersof an uncontrolled, politically corrupted, or irresponsible bureau-cracy; the corruptibility of legislative processes; the imprecision ofpopular control of administration; and judicial and executive insti-tutions that can balance capacity with control in a constitutionallyappropriate manner.

(Bertelli and Lynn 2006: 43)

The field could now identify itself with a literature evincing a subtle, pragmatic wisdom concerning administrative practice that was, at the sametime, infused with democratic idealism and a keen awareness, if not a com-plete conceptual grasp, of the challenges created by America’s constitutionalseparation of powers.

The “New (Public) Governance”

In the early 1990s, many of public administration’s leading scholars and practitioners began, once again but this time on an international scale, toproclaim “the new”, which was simultaneously available in two distinct brands:NPM, which emphasized “markets”, and the new (public) governance, whichemphasized “networks”. The story of NPM has been often told and, as arguedearlier, is of less interest than the new (public) governance, which remainsascendant and possibly transformative.

Dutch, American and British experts separately asserted the propositionthat advanced democracies are moving toward new forms of governance;indeed, some suggested “governance without government”, a mode of societalsteering that has been de-bureaucratized in favor of emergent hybridized and consociational institutions of civil society. These emergent discourses were necessarily influenced by national and regional institutional historiesand contexts; Anglophone, American, Nordic and Continental ideas had their

own flavors (Klijn 2008). Common themes and emphases soon became evident, however.

New governance in Europe

Introducing his 1993 edited symposium volume Modern Governance: NewGovernment–Society Interactions, Jan Kooiman discussed “new patterns ofinteraction between government and society” and “new ways of governingand governance” that he termed “social-political governance” (1993: 2, 3).In response to the increasing complexity, dynamics and diversity of societies,the public and private sectors are, he said, engaging in continuous inter-actions and acting in conjunction with one another rather than separately.The result has been new structural forms of interactive steering, managing,controlling and guiding in certain sectors.

R. A. W. Rhodes, writing in 1996 of the changes he perceived in Britishgovernment, redefined, in the manner of Gaus in the early 1920s, the hereto-fore generic and descriptive term “governance” as signifying “a change inthe meaning of government, referring to a new process of governing; or achanged condition of ordered rule; or the new method by which society isgoverned” (Rhodes 1996: 652–3). Specifically, Rhodes argued, to marketsand hierarchies (bureaucracies) as governing structures can now be addednetworks characterized by trust and mutual adjustment. Later Mark Bevir,Rhodes and Patrick Weller, organizers of a symposium similar to Kooiman’sa decade before, used “governance” to encapsulate “the changing form androle of the state in advanced industrial societies”, especially focusing on public-sector reform and on “the changing boundary between state and civilsociety” (2003: 13).

Subsequently, Stephen Osborne proposed that we are entering a new stagein the profession of public administration and management. The first stage,traditional public administration, while emphasizing politics, gave short shriftto management. The next stage, NPM, focused on management while relegat-ing democratic institutions to the status of mere context if not of outrightimpediment. From an intellectual perspective, New Public Governance (NPG)corrects the theoretical and practical shortcomings of its predecessors, encom-passing the contemporary complexities and realities of governing by drawingon organizational sociology and network theory rather than on political science or public-choice economics in order to overcome the fragmentationand uncoordinated character of twentieth-century managerial practice.

The “new (public) governance” has been proposed elsewhere in theAnglophone world. In Canada, Gilles Paquet spoke of “a shift from a some-what ‘defensive,’ anti-democratic, centralizing, homogenizing and hierarchicalformer regime ruled by elites toward a more ‘open,’ communitarian, non-centralizing, pluralistic and distributed governance regime” (1999: 73). OfAustralian social policy, Tim Reddel says that “a new governance discourse

offers some promise as a policy framework that can re-conceptualize thestate–community (and market) relationship and deliver improved com-munity outcomes, particularly in the context of place-based or spatial policiesand programs” and featuring “dialogue, deliberation, and association” and“engagement with a strong civil society” (Reddel 2002: 50–1, 54, 60).

In his 2008 assessment of the European literature on governance, Erik-Hans Klijn concluded that the term “governance”, in so far as it identifiesa departure from traditional emphasis on the internal functioning of govern-ments, refers to governance networks, that is, to government’s “relationshipswith other actors and the process of handling complex decisions and imple-mentation processes” (Klijn 2008: 510–11). “Governance”, he says, “is theprocess that takes place within governance networks”, that is, within “a web of relationships between government, business and civil society actors[which are] based on interdependencies [that] are not necessarily equitable”(p. 511). Thus, in contrast to those, such as Rhodes and Kooiman, who haveviewed the new governance as traditional structures plus networks, Klijn viewsit as networks alone.

New governance in America

In 1994, American scholars associated with the National Academy of PublicAdministration identified a “cluster of ideas and symbols” they termed “newgovernance” (Dewitt et al. 1994). Inspired by books such as David Osborneand Ted Gaebler’s Reinventing Government (1992) and Michael Barzelay andBabak Armajani’s Breaking Through Bureaucracy (1992), “new governance”was said to encompass “[n]ew ways of doing the public’s business”, drivenprimarily by policy specialists and already being experimentally introducedby state and local governments, and soon to be attempted at the federal level (by the Clinton administration). “New governance” anticipated and advocated further changes in the participants, purposes, means and politicsof government. Participants, they believed, would come to include alliancesand collaboration in lieu of fragmentation and decentralization of decision-making. Citizens would be empowered rather than being merely the passiveobjects of government initiatives.

In 2000, American political scientist Lester Salamon defined the “new governance” as having two distinctive features: its collaborative nature and its recognition that this form of governance involves challenges as wellas opportunities. As Salamon sees it, each of a “dizzying array” of new administrative forms or tools had its own skill requirements, its own deliverymechanisms, and its own political economy (2000: 1612, 1613). Most of the newtools are highly “indirect”, involving the sharing with non-governmental actorsof discretion over the exercise of public authority and the allocation of public resources. His initial term for this phenomenon – he began writingabout it in 1980 – was “third party government”. In support of his claims,

he quotes Donald Kettl as having observed that “[e]very major policy ini-tiative launched by the federal government since World War II . . . has beenmanaged through public–private partnerships” (Salamon 2000: 1615, citingKettl 1993: 4). A similar phenomenon, under such rubrics as “pillarization”and “subsidiarity”, has, he notes, been occurring in Western Europe.

A new governance discourse has arisen in specific policy domains as well.Robert Durant argues that traditional command-and-control regulation isinadequate for non-point source pollutants such as greenhouse gases, andthat regulatory management concerning environmental and natural-resourcepolicies must reconnect with stakeholders and give greater flexibility to bothregulators and regulated communities (Durant 2004). American public adminis-tration scholars have viewed national and state rural development councilsas an example of a new governance featuring reduced federal direction infavor of intergovernmental collaboration and increased involvement ofstate, local and private agencies (Radin 1996). Americans Lisa Bingham, TinaNabatchi and Rosemary O’Leary (2005) discuss a “different face” of the “newgovernance”, which will be the “watchword for the next millennium”.

Practitioners are using new quasi-legislative and quasi-judicial govern-ance processes, including deliberative democracy, e-democracy,public conversations, participatory budgeting, citizen juries, studycircles, collaborative policy making, and alternative dispute resolu-tion, to permit citizens and stakeholders to actively participate inthe work of government.

p. 547

The stories compared

Both of the stories sketched above feature claims about a changed world:the institutional status quo has been, or is being, replaced by something sodifferent that it constitutes a new way of thinking and addressing public issues.However, a convincing argument of any kind, but certainly a scholarly one,requires not only a claim concerning what is (probably) true but also fouradditional elements: reasons, evidence, a warrant (authorizations or justi-fications based on theory or principle) and, finally, qualifications (limits ofapplicability).3 What are the arguments central to each of the two stories?Are they equally convincing? If not, why not? If so, what are the implica-tions for thought and practice?

The arguments offered on behalf of both the new (public) administrationand the new (public) governance generally include the required elements. Their respective claims are easy to identify: “a new (public) administrationbased on new government institutions has emerged over the last thirtyyears” in the first story, and “new modes of governing involving networksand the sharing of authority with civil society institutions are replacing

traditional reliance on public authority” in the second. Reasons are similarin both arguments and take the form: “traditional institutions proved to beinadequate for addressing problems of increasing complexity associatedwith changes in society, technology, and the economy”.

When it comes to the evidence and the warrants supporting the claims andreasons, however, the quality of the two arguments diverges. One reason may be that new governance ideas are still being formulated and contested.Another reason goes directly to the quality of the arguments being put forward for the new (public) governance: their normative, theoretical andempirical dimensions are not always clearly distinguished, and are oftenconflated. While disentangling these strands has not been attempted here, itis important to identify problems of evidence, theory and qualifications.

The problem of evidence

During the emergence of the American administrative state, academic publicadministration and public administration as codified professional practice werelargely inchoate and emergent. The political scientists, municipal research-bureau staff, and activist reformers who created both the reality and the historical record of institutional transformation were themselves partici-pants in the historical process. As a result, evidence was predominantly basedon experience, direct observation and official data. When first articulated,the new public administration was regarded as a fait accompli; its examplesconstituted evidence of a new reality, a new paradigm in the literal meaningof that term.

Now, given the existence of a mature profession, so-called cycles ofadministrative reform can run their course largely in the academic imagina-tion and in the specialized precincts of technocrats and consultancies. Thus,examples of a new (public) governance are often carefully selected harbingersof a reality possibly in the making rather than constituting a balancedassessment of the range of actual changes in state–society relations. Manycontributors to the new governance literature offer broad generalizations concerning institutional change or examples of new forms of goverance with-out saying much about whether habits of thought among policy-makers havechanged. Moreover, contending as it is with still other claims on behalf of“the new”, such as the “new city management” (Hambleton 2002; Hambletonand Gross 2007) and the “new public service” (Denhardt and Denhardt 2002,2003), in which there is as much advocacy as analysis, it is not yet clear whetherthe new (public) governance reflects just another cycle of intellectual fashionstimulated as much by who desires reform as by intellectual curiosity or theemergence of what will eventually be seen as a self-evident new reality basedon new habits of thought, the evidence for which is everywhere at hand.

However, the difference in the evidentiary bases for the two stories and their claims should not be overstated. Of the established American

administrative state, White noted that “[w]ithout any adequate proof of thesuperiority of the new institutions at hand, it is clear that they have by andlarge commended themselves to the voters. As soon as adequate measuresof success or failure are at hand, their contribution may be made more clear” (1933: 236). The new (public) governance literature, moreover, featuresincreasing empirical depth. Salamon’s claims concerning the new reality ofthird-party government, for example, are supported by data on the expand-ing role and size of the nonprofit sector in publicly financed service deliveryand also by a growing body of official data and scientific research. The literature on the new governance in the European Union and on new formsof citizen participation in governance features a good deal of experientialreflection and some scientific investigation. Research such as that of Radin(1996) and Durant (2004) use the new governance as a frame for analyzinginstitutional change in rural economic development and in environmental andnatural resource management respectively. Studies such as that of Taylor (2002)demonstrate the extent of government influence in new governance structures.Empirical knowledge is accumulating.

In summary, empirical evidence supporting claims that a new (public) governance is redrawing state–society boundaries is contested and, in qualityand extent, as yet unconvincing.

The problem of theory

Proclaiming “the new” in both stories depends for intellectual leverage andsalience on a more principled warrant, that is, on a theory that justifies thelogic of linking claims, reasons and evidence.

Claims on behalf of the new (public) administration were thought to bevindicated by the needs of the polity in a modernizing and industrializingsociety as expressed through America’s constitutional scheme of governance.The warrant can be stated more generally. The cumulative and combinedeffects of several historical developments over three centuries – the advent ofthe nation state, the transformation of mercantilism into capitalism, demo-cratic and specifically republican revolutions, industrialization, modernizationand urbanization, and international conflicts – resulted in the emergence in the developed world of centralized, hierarchically organized, politicallyaccountable administrative states founded on the rule of law (Aucoin 1997;Dahl and Lindblom 1953: 511; du Gay 2000; Goodsell 1983; Meier 1997;Peters and Pierre 2003).

In contrast, most arguments for a new (public) governance are primarilyanalytical rather than theoretical, featuring classification and generalizationrather than a causal logic rooted in democratic principles. In many of thesearguments, a necessarily selective body of evidence is, in effect, assumed tospeak for itself. The fact that numerous countries are making reforms with afamily resemblance to one another is often presumed to suggest inevitability

and a global convergence of governing paradigms (cf. arguments for NPM).Advocates for new paradigms might be forgiven for omitting such theorizing;their goal is an immediate and practical one: to alter the thinking of policy-makers, stakeholders and administrators in support of new, arguably betterforms of governance. Such omissions are less forgivable in scholarly discourse,however, which is, and ought to be, held to more rigorous standards for claimsand for the logic of the reasons and evidence that support them.

What might be the warrant for the argument that a new (public) govern-ance is, within a generation, replacing centuries-old institutional expressionsof republican rule? Why might it be in the interests of the stakeholders of representative democracy – citizens, elected representatives, and legal institutions – to dispense with bureaucracy and the types of account-ability mechanisms associated with it in favor of more dispersed, diffuse and ambiguous accountability? And why now?

The most popular warrant is the vague notion of “globalization”. The triumph of more aggressive forms of capitalism (at least until the global financial crises beginning in 2008), of border-erasing technologies, and of competition-based, performance-oriented social allocation, that is, of the global marketplace, over socialism, state-directed social allocation, and socialdemocratic welfare statism was, it is asserted, bound to affect all states insimilar ways. Globalization, in other words, represents a disruptive his-torical discontinuity, the significance of which for state–society relations isas great as the emergence of the nation-state, of market capitalism and ofrepresentative democracy.

Fred Riggs and Judith Merkle have articulated the basis for a more prin-cipled warrant for the new institutional developments. The basic principleof democracy as an aspect of modernity, in Riggs’s account, involved thereplacement of top-down monarchic authority with bottom-up representa-tion: dominated subjects were replaced by free citizens able to participate in governance, choose their governors, and hold them accountable throughperiodic elections. But, he says, “[i]t has never been easy in even the mostdemocratic countries for the organs of representative government to sustaineffective control over their bureaucracies” (1997: 350), and it became moredifficult as those bureaucracies were being rationalized in order that they mightapproach industrial reliability and efficiency.

Merkle offers a similar type of reasoning (1980). In her account, futurepresident Woodrow Wilson initiated a new idea: “power wielded without regardto persons [either monarch or the people] but rather to attain social endswhose usefulness and means of attainment could eventually be determinedby science itself ” (1980: 280). She continues: “It was a theory of politics thatproclaimed the victory of the experts and defined ‘politics,’ or the contestfor power, as obsolete” (p. 281). An implication of Merkle’s reasoning is thatall the pathologies of bureaucracy that were coming to be widely recognizedwill influence the premises of political choice.

All of these modes of influence [by bureaucracies] tend to break downthe ability of the political system to control bureaucracy, while at thesame time they introduce the values and the vested interests of thestate’s own administrative “servant” into the political process itself.

(Merkle 1980: 282; cf. Karl 1976)

Attempts at democratic control, as Riggs also notes, only undermine the effectiveness of administrative action, ensuring a fundamental tension betweenthem.4

Based on these types of reasoning, the warrant for a new (public) govern-ance might be that civil society institutions, governing elites and other stakeholders are coming to the view that the inevitable tensions between representative democracy and nonrepresentative bureaucracy should be ameli-orated by redrawing the state–society boundary so as to enlarge the role ofcivil society – a political preference which will come to be accepted by policy-makers and judges as consistent with republican principles and therefore legitimate.

Of interest at this point is an integrative concept originating in socio-logical theory, that of an “organizational field”, that is, a “recognized area of institutional life . . . involving the totality of relevant actors” (DiMaggioand Powell 1991: 64–5).5 Paul DiMaggio and Walter Powell argue that “[b]ure-aucratization and other forms of organizational change occur as the resultof processes that make organizations more similar without necessarily makingthem more efficient”. Three forces that induce isomorphism, that is, conver-gent structural /functional arrangements, are coercive, i.e. political influence(operating as an exogenous influence on organizations) that ensures legiti-macy; mimetic, i.e. standardization of responses to uncertainty; and normative,i.e. standardization reflecting professional beliefs and values (DiMaggio andPowell 1991: 67). In other words, choices of administrative arrangements maybe a resultant of socialized rather than rational choice within organizationalfields, of imitation, coercion, or ideology rather than of calculation and analysis. We should not be surprised to find convergence on a functional orspecific policy institution even if not on a whole-of-government scale.

Elaborating on this line of reasoning is Powell’s (1990) elucidation of theconditions that give rise to network forms of organization. Under certaincircumstances, he argues, exchange relationships may be governed by recipro-city and collaboration rather than by (complete, incomplete, or implicit) con-tracts or structures of formal authority. In general, says Powell (1990: 326–7),

networks appear to involve a distinctive combination of factors –skilled labor, some degree of employment security, salaries ratherthan piece rates, some externally-provided mechanisms for jobtraining, relative equity among the participants, a legal system withrelaxed antitrust standards, and national policies that promote research

and development . . . which seldom exist in sufficient measure with-out a political and legal infrastructure to support them.

Cross-national variation in the frequency of network forms may be explained,Powell suggests, by variations in state policies that support and sustain collaborations. Differentiating among administrative forms in this way mayhelp in identifying the underlying dynamics of change.

Finally, R. L. Jepperson and J. W. Meyer, also sociologists, adduce a frame-work of explanation that views the evolution of the formal organization ofsociety as governed by the modern polity, by which they mean

the system of rules conferring societal authority in pursuit of collec-tive ends, establishing agents of collective regulation and intervention.. . . We have in mind processes such as monetarization and demo-cratization – the construction of markets and rights – and the institutionalization of goals such as collective progress and justice.

(Jepperson and Meyer 1991: 206)

Their framework accommodates global influences.

Our line of argument proposes that a wider polity (often worldwide)of universalistic collective definitions plays a governing role, com-bined with an expanding set of subunit national societies competingand copying each other within this frame.

(1991: 209)

These authors, then, suggest the possibility of a mimetic isomorphism thatallows for degrees of convergence on a new (public) governance that are consistent with path dependence.

The new skepticism

The earlier American story attracted little fundamental criticism; virtuallyno one argued that a new administrative state was not in fact emerging,although many noted in qualification that change was proceeding gradu-ally. But qualifications and outright skepticism abound in the story of new(public) governance. Johan Olsen (2006: 2) quoted Hood (1996: 268) thatmany reform proposals are “repackaged versions of ideas that have been inpublic administration since its beginnings” and that “new” approaches fre-quently rehash old ideas (Kettl 1993: 408). Olsen also notes that Rhodes (1994;see also Davis and Rhodes 2000) predicted a return to bureaucracy ten yearsago, and Peters (1999: 104–5) sees “a possible return to Weber’s organiza-tional archetypes”. Based on comparative analysis, Pollitt and Bouckaert say that “The idea of a single, and now totally obsolete, ancient regime is

as implausible as the suggestion that there is now a global recipe which willreliably ‘reinvent’ governments” (2004: 63). Even Kooiman, a progenitor ofthe governance perspective, concedes that “the state is still very much alive”(Kooiman 2003: 130).

Of particular interest is the analysis of governance reforms in Anglophoneand European states by Bevir, Rhodes and Weller. They ask: “What is theplot of our story?” The complex plot comprises twelve points. Notable amongthem is the following: “the beliefs and practices of elite actors [engaged inreform] originate in the traditions they have inherited. They construct issuesor dilemmas out of experiences infused with these traditions” (2003: 202).Further, “[g]overnance is constructed differently and continuously reconstructedso there can be no one set of tools” (p. 203). Moreover, specific reforms havedifferent meanings in different countries; new governance in the Netherlandswill differ from new governance in the UK. There is “no universal processof globalization driving public sector reform” (p. 203). The primary elites,moreover, remain central agencies. Thus, the new (public) governance is, atbest, a highly nuanced and qualified story.

New (Public) Governance: a new paradigm?

Comparing a completed story with a work in progress is problematic. Thereis value in the effort, however, if the question is posed as follows: Might thestory of new (public) governance have the same ending as the story of thenew (public) administration? That is, might the qualifier “new” eventuallybe dropped and “public governance” replace “public administration” as thegeneral term that encompasses all forms of state–society interactions that westudy and practice?

A deeper issue lurks beneath this question, however: whether the bases of legitimacy for state-sponsored action will shift decisively toward more open and transparent civil society institutions. Ultimately this is a politicalquestion. The discussion in the preceding section sketched one political pathtoward such a result: pressure on elected officials from citizens disaffectedby strong-state traditions that persuades them to advocate and authorize altogether different forms of societal steering, as did indeed take place inAmerica in the decades from 1890 to 1930.

However, skepticism concerning such a political development is warranted.Any paradigm which requires political actors to change fundamentally theirways of viewing their roles without concomitant changes in constitutions andpolitical institutions is arguably doomed to become just another academicfashion that will run its course. Of Salamon’s approach to new governance,for example, Guy Peters has argued that there are more fundamental considerations than “tools of governance” at work in American politics. Indeed, there are more fundamental issues at work in democratic politics worldwide.

One of those fundamental considerations is constitutional governance. With all that has been new over centuries of state development, what hasendured has been those institutions that guarantee the legitimacy of delegatedauthority: legislatures and judicial bodies along with norms of public service (Lynn 2006). In America, and in other countries as well, to the extent that electoral politics involves the direct engagement of interest groups and otherconstituencies, elected officials will continue to be under pressure, when thelegitimacy of the uses of delegated authority is called into question, toenforce traditional modes and concepts of accountability. If there was any doubt about the truth of this proposition, the responses of governmentsworldwide to the financial and economic crises that engulfed the world beginning in 2008 should have dispelled it.

Constraining managerial tools and practices, and in a sense transcendingthem, then, are the law and legal institutions and their derivative checks and balances. “State power is the great antagonist against which the rule oflaw must forever be addressed,” says H. W. Jones (1958: 144). As AnthonyBertelli notes, administrative law regimes exist to police the delegation of powers from sovereign authority to bureaus. “The commitment touphold administrative law is made credible in most of the world by a thirdparty[, notably] the courts” or by other arrangements, such as ombudsmenor inspectors-general or other entities possessing independent authority to review and report (Bertelli 2005: 151). Note Walter Kickert and JanHakvoort (2000: 251), the “all-pervasive dominance of administrative law,combined with the rational, deductive, comprehensive, systematic way of legalthinking, must have serious implications for the form of ‘public manage-ment’ that is adopted” in Continental countries dominated by civil law andin other countries as well.

Constitutions and constitutional institutions will continue to frame the evolution of states and, as a consequence, their managerial institutions. Theoverthrow of current constitutional arrangements is not in prospect. Withinthese institutional frameworks, institutions will continue to evolve, and we shall see, perhaps even anticipate, patterns of interaction that are new. But the narrative of differentiation that emerges from comparative researchpresages distinctive futures, not convergence, for national administrations.These futures may, over time, come to look different from the contemporaryreality of public governance, but only in so far as the central problematic of democratic governance, accountability to representative institutions, isresolved in ways that sustain the constitutional legitimacy of state action.Whatever may be new will be rooted in soil that is very old indeed.

1 American public administration scholar Nicholas Henry (2007) has identified sixparadigms in the field’s first century, the last of them being governance.

2 He noted, however, that the new administration had been anticipated by Britishessayist Walter Bagehot, whose admiring biographies of Victorian administratorsprefigured the challenges of the modern administrative state.

3 This paragraph is based on Toulmin (2003) as interpreted by Booth, Colomb, andWilliams (2003).

4 Argues Riggs (1997: 347), “No bureaucracies, modern or traditional, are demo-cratic; they are instead administrative and hierarchic.” Of the emergence of theAmerican administrative state, Barry Karl (1976: 503) notes that “the profession-alization of administration created national interests among administrative specialiststhemselves, separating them from the local attachments their [Jacksonian] pre-decessors had understood so well,” thus deepening the problems of democraticaccountability. The consequences of professionalization may be even more evidentin European democracies.

5 This part of the discussion is adapted from Lynn (2006).

References

Aucoin, Peter (1997) “The Design of Public organizations for the 21st Century: WhyBureaucracy Will Survive in Public Management”, Canadian Public Administration40 (2): 290–306.

Bertelli, Anthony M. (2005) “Law and Public Administration”, in Ewan Ferlie, LaurenceE. Lynn, Jr and Christopher Pollitt (eds) The Oxford Handbook of Public Man-agement, Oxford: Oxford University Press.

Bertelli, Anthony M. and Laurence E. Lynn, Jr (2006) Madison’s Managers: PublicAdministration and the Constitution, Baltimore, Md: Johns Hopkins University Press.

Bevir, Mark, Rhodes, R. A. W. and Weller, Patrick (2003) “Comparative Govern-ance: Prospects and Lessons”, Public Administration, 81 (1): 191–210.

Bingham, Lisa Blomgren, Nabatchi, Tina and O’Leary, Rosemary (2005) “The NewGovernance: Practices and Processes for Stakeholder and Citizen Participation inthe Work of Government”, Public Administration Review, 65 (5): 547–58.

Booth, Wayne C., Colomb, Gregory G. and Williams, Joseph M. (2003) The Craftof Research, 2nd edn, Chicago, Ill.: University of Chicago Press.

Cleveland, Frederick A. (1913) Organized Democracy, New York: Longmans,Green.

Dahl, Robert A., and Charles E. Lindblom (1953) Politics, Economics, and Welfare:Planning and Politico-economic Systems Resolved into Basic Social Processes, NewYork: Harper Torchbooks.

Davis, G., and Rhodes, R. A. W. (2000) “From Hierarchy to Contracts and BackAgain: Reforming the Australian Public Service.” In Institutions on the edge:Capacity for governance, ed. M. Keeting, J. Wanna, and P. Weller, 74–98. St.Leonards, Australia: Allen and Unwin.

Denhardt, Robert B. and Denhardt, Janet Vinzant (2002) “The New Public Service:Serving Rather than Steering”, Public Administration Review, 60 (6): 549–59.

Denhardt, Robert B. and Denhardt, Janet Vinzant (2003) The New Public Service:Serving, Not Steering, Armonk, NY: M. E. Sharpe.

DeWitt, John, Kettl, Donald F., Dyer, Barbara and Lovan, W. Robert (1994)“What Will New Governance Mean for the Federal Government?”, PublicAdministration Review, 54 (2): 170–6.

DiMaggio, Paul J. and Powell, W. W. (1991) “The Iron Cage Revisited: InstitutionalIsomorphism and Collective Rationality in Organization Fields, in W. W. Powelland Paul J. DiMaggio (eds), The New Institutionalism in Organizational Analysis.Chicago, IL: The University of Chicago Press.

Downs, Anthony (1972) “Up and Down with Ecology: The Issue-Attention Cycle”,The Public Interest, 28: 38–50.

du Gay, Paul (2000) “In praise of Bureaucracy: Weber–Organization–Ethics”,Milton Keynes, UK: Open University Press.

Dunleavy, Patrick, Margetts, Helen, Bastow, Simon and Tinkler, Jane (2006) “NewPublic Management Is Dead – Long Live Digital-era Governance”, Journal of PublicAdministration Research and Theory, 16: 467–94.

Durant, Robert F. (2004) “Toward a New Governance Paradigm for Environmentaland Natural Resources Management in the 21st Century?”, Administration andSociety, 35 (6): 643–82.

Gaus, John (1923/24) “The New Problem of Administration”, Minnesota LawReview, 8: 217–31.

Gaus, John M., Leonard D. White, and Marshall E. Dimock, eds. (1936) TheFrontiers of Public Administration. Chicago: University of Chicago Press.

Goodsell, Charles T. (1983) The Case for Bureaucracy: A Public Administration Polemic2d ed., Chatham, NJ: Chatham House Publishers.

Hambleton, R. (2002) “The New City Management”, in R. Hambleton, H. V.Savitch and M. Stewart (eds) Globalism and Local Democracy: Challenge and Changein Europe and North America, Basingstoke /New York: Palgrave Macmillan.

Hambleton R. and Gross J. S. (eds) (2007) Governing Cities in a Global Era: UrbanInnovation, Competition and Democratic Reform, Basingstoke/New York: PalgraveMacmillan.

Henry, Nicholas (2007) Public Administration and Public Affairs, 10th edn, UpperSaddle River, NJ: Pearson Prentice-Hall.

Hood, Christopher (1996) “Beyond ‘Progressivism’: A New ‘global paradigm’ in public management?” International Journal of Public Administration, 19: 151–177.

Jepperson, R. L. and Meyer, J. W. (1991) “The Public Order and the Constructionof Formal Organizations”, in W. W. Powell and P. J. DiMaggio (eds), The New Institutionalism in Organizational Analysis, Chicago, IL: The University ofChicago Press.

Jones, H. W. (1958) “The Rule of Law and the Welfare State”, Columbia Law Review,58: 143–56.

Karl, Barry (1976) “Public Administration and American History: A Century ofProfessionalism”, Public Administration Review, 36: 489–503.

Kettl, Donald F. (1993) Sharing Power: Public Governance and Private Markets,Washington, DC: The Brookings Institution.

Kickert, Walter J. M. and Hakvoort, Jan L. M. (2000) “Public Governance in Europe:A Historical–Institutional Tour d’Horizon”, in Oscar Van Heflin, Walter J. M.Kickert and Jacques J. A. Thomassen (eds) Governance in Modern Society: Effects,Change, and Formation of Government Institutions, Dordrecht: Kluwer.

Klijn, Erik-Hans (2008) “Governance and Governance Networks in Europe: AnAssessment of Ten Years of Research on the Theme”, Public Management Review,10: 505–25.

Kooiman, Jan (ed.) (1993) Modern Governance: New Government–Society Inter-actions, London: Sage.

Kooiman, Jan (2003) Governing as Governance, London: Sage.Laski, Harold J. (1923) “The Growth of Administrative Discretion”, Journal of Public

Administration, 1: 92–100.Lynn, Laurence E., Jr. (2006) Public Management: Old and New, London/New

York: Routledge.Margolis, Howard (1993) Paradigms and Barriers: How Habits of Mind Govern

Scientific Beliefs, Chicago, Ill.: University of Chicago Press.Meier, K. J. (1997) “Bureaucracy and Democracy: The Case for More Bureaucracy

and Less Democracy”, Public Administration Review, 57 (3): 193–99.Merkle, Judith A. (1980) Management and Ideology: The Legacy of the Inter-

national Scientific Management Movement, Berkeley, CA: University of CaliforniaPress.

Merriam, Charles E. (1940) “Public Administration and Political Theory”, Journalof Social Philosophy, 5: 293–308.

North, Douglass C. (1990) Institutions, Institutional Change, and Economic Perform-ance, New York: Cambridge University Press.

Olsen, Johan P. (2006) “Maybe It Is Time to Rediscover Bureaucracy,” Journal ofPublic Administration Research and Theory, 16: 1–24.

Paquet, Gilles (1999) “Innovations in Governance in Canada”, Optimum, 29 (2–3):71–81.

Peters, B. Guy (1999) Institutional Theory in Political Science, London and New York:Pinter.

Peters, B. Guy and Jon Pierre (2003) “Introduction: The role of Public Administra-tion in Governing”, in Handbook of public administration, ed. B. Guy Peters andJon Pierre, 1–9, London: Sage.

Pfiffner, John M. (1935) Public Administration, New York: Ronald.Pollitt, Christopher and Geert Bouckaert (2004) Public Management Reform: A

Comparative Analysis. 2d ed., Oxford: Oxford University Press.Powell, W. W. (1990) “Neither Market nor Hierarchy: Network Forms of Organ-

ization”, Research in Organizational Behavior, 12: 295–336.Radin, Beryl (ed.) (1996) New Governance for Rural America: Creating Inter-

governmental Partnerships, Lawrence, Kan.: University Press of Kansas.Reddel, Tim (2002). “Beyond Participation, Hierarchies, Management and Markets:

‘New’ Governance and Place Policies”, Australian Journal of Public Administration,61 (1): 50–63.

Rhodes, R. A. W. (1994) “The Hollowing-out of the State”, Political Quarterly, 65:138–51.

Rhodes, R. A. W. (1996) “The New Governance: Governing without Government”,Political Studies, 44 (4): 652–67.

Rhodes, R. A. W. (2000) “Governance and Public Administration”, in J. Pierre (ed.),Debating Governance. Oxford: Oxford University Press, pp. 54–90.

Riggs, Fred W. (1997) “Modernity and Bureaucracy”, Public Administration Review,57: 347–53.

Salamon, Lester M. (2000) “The New Governance and Tools of GovernmentAction: An Introduction”, Fordham Urban Law Journal, 28: 1611–74.

Taylor, Andrew (2002) “ ‘Arm’s Length but Hands On’: Mapping the New Govern-ance. The Department of National Heritage and Cultural Politics in Britain”, PublicAdministration, 75 (3): 441–66.

Toulmin, Stephen Edelston (2003) The Uses of Argument, Cambridge: CambridgeUniversity Press.

Weimer, David L. (1995) “Institutional Design: Overview”, in David L. Weimer (ed.)Institutional Design, Boston, Mass.: Kluwer.

White, Leonard D. (1933) Trends in Public Administration, New York: McGraw-Hill.White, Leonard D. (1955) Introduction to the Study of Public Administration, 4th edn,

New York: The Macmillan Company.

Part II

GOVERNANCE ANDINTERORGANIZATIONAL

PARTNERSHIPS TO DELIVERPUBLIC SERVICES

THEORY OF ORGANIZATIONALPARTNERSHIPS: PARTNERSHIP

ADVANTAGES, DISADVANTAGESAND SUCCESS FACTORS

Ronald W. McQuaid

Promoting “partnership” and greater interagency cooperation between govern-ment departments, public agencies, private companies and the third sectorhas become a staple of strategies to promote social and labor-market inclusionat national and supranational levels, for instance internationally (CEC 2003,2001: 6; OECD 2008) and the UK (DWP 2006). Area-based strategies to tackle social and labor-market exclusion have particularly used partnershipapproaches involving different organizations and forms of relationship – forthe UK government, “renewal relies on local communities”, and non-publicbodies have a leading role to play in promoting employability, regenerationand inclusion (SEU 2001; McQuaid and Lindsay 2005; McQuaid et al. 2007).

Across and beyond the EU, policy-makers are turning to new forms ofpartnership and seeking to include a wider range of stakeholders in the design,planning and delivery of policies. In the case of employment policies, thisreflects an acceptance that, in order for employability policies or interventionsto address the range of complex and multi-dimensional problems faced by unemployed and economically inactive people, multi-agency approachesare required. For instance, as governments refocused their welfare-to-workstrategies on those claiming long-term income-based benefits in many Euro-pean countries (Lindsay and McQuaid 2008, 2009) and incapacity benefitsin the UK (Lindsay et al. 2008), they also “opened up” employability servicesto a wider range of stakeholders, in an attempt to extend their quality andreach, and to access specialist knowledge and expertise.

Different types or organization of partnerships are appropriate in differ-ent circumstances, and a key strategic issue is to identify and choose an appropriate type. Some of the main dimensions of partnership are: what thepartnership is seeking to do, i.e. its purpose and whether it is strategic orproject-driven; who is involved, i.e. the key actors and the structure of their

relationship in the partnership; when, i.e. the timing or stage of developmentof the partnership process and changing relationships and activities over time;where, i.e. the spatial dimension; and how the activities are carried out, i.e.the implementation mechanisms (McQuaid 2000).

This chapter considers a number of conceptual and policy issues surround-ing partnership working, including interagency cooperation, mainly using examples in the areas of employability and local regeneration policy.Following this introduction the chapter addresses: definitions of “partner-ship”; potential benefits associated with partnerships/interagency coopera-tion; limitations and problems with partnerships/interagency cooperation;critical success factors in effective partnerships; some implications for governance; and conclusions.

Definitions of partnership

The term “partnership” covers a multi-dimensional continuum of widely differing concepts and practices, and is used to describe a variety of typesof relationship in a myriad of circumstances and locations. Some, such aspublic–private partnerships, often refer to specific forms of contractual relationships, although this chapter is concerned with wider formal and informal interagency cooperation. Such is the rhetorical power of the lan-guage of “partnership” that concerns have been raised that the concept hasbecome little more than a buzzword to “sprinkle liberally through fundingapplications” (Osborne 1998); or an idea so ubiquitous in major policy initiatives that it defies definition (Rowe and Devanney 2003) and risks losing its analytical value (Miller 1999). Lankshear et al. (1997: 88–9) suggestthat key terms like “partnership” are

words that cross discursive boundaries, spanning multiple world-views,interests and value systems. They all carry positive connotations and name ideals to which people who embrace different – and oftenincompatible – aspirations, purposes, interests and investmentsclaim allegiance.

The concept of partnership in service delivery arose, during the 1980s and1990s, as a canon of public policy and private enterprise (Knox 2002). TheOECD (1990: 18) has defined partnerships as:

Systems of formalised co-operation, grounded in legally bindingarrangements or informal understandings, co-operative workingrelationships, and mutually adopted plans among a number of insti-tutions. They involve agreements on policy and programme objectivesand the sharing of responsibility, resources, risks and benefits overa specified period of time.

GOVERNANCE AND INTERORGANIZATIONAL PARTNERSHIPS

Further to this very broad definition, a number of more context-specificdefinitions of partnership have been articulated. Harding (1990: 110) sets outa general definition of “private–public partnership” as “any action which relieson the agreement of actors in the public and private sectors and which alsocontributes in some way to improving the urban economy and the qualityof life”. Bailey et al.’s (1995: 293) more specific definition of partnerships forurban regeneration speaks of “the mobilisation of a coalition of interests drawnfrom more than one sector in order to prepare and oversee an agreed strategyfor regeneration of a defined area”.

It is useful to distinguish partnerships at different levels of organization,such as at strategic or project levels, or geographically at regional and locallevels. Snape and Stewart (1996) are interested in different levels of inter-agency cooperation – they distinguish between three ideal-typical forms of partnership working in social-inclusion policy: facilitating partnerships, which manage longstanding strategic policy issues; coordinating partnerships,which are concerned with the management and implementation of policy based on broadly agreed priorities; and implementing partnerships, which arepragmatic and concerned with specific, mutually beneficial projects. To this can be added more strategic-goal agreement partnerships which seek toidentify key directions and aims, and how these might be achieved throughpartnership working.

Drawing on a number of existing approaches, Hutchinson and Campbell(1998: 9) suggest that there is a consensus around a number of defining features: partnerships bring together a coalition of interests drawn from more than one sector to generate agreement; partnerships have common aimsand a strategy to achieve them; partnerships share risks, resources andskills; partnerships achieve mutual benefit and synergy.

This discussion demonstrates that partnership remains a varied and ambi-guous concept. In the UK context, the debate has been further complicatedby the government’s application of the language of partnership to programsand relationships that in fact involve the allocation of resources on the basisof competitive tendering to provide services. These include public–private partnerships (and private finance initiatives, or PFIs, in the UK) (McQuaidand Scheerer 2010). Government departments and funders are required to fulfill the dual role of acting as strategic partners, working with other government agencies and stakeholders to shape the general framework forlocal policy implementation, while also acting as a funder, contracting outservices often through some of the same stakeholders.

Increasingly important are partnerships between different public-sector bodies which link different types of services (e.g. the linking of welfare allow-ances, health, and job search, etc.). There may be tensions between organizingactivities along specific functions and tasks (e.g. providing welfare benefitsin a cost-effective way) and more client-centered approaches where the fullrange of issues facing a client need to be dealt with in order to help them to

THEORY OF ORGANIZATIONAL PARTNERSHIPS

move on (e.g. to employment). For example, it may be important to train a person to manage their health condition first, and to remove their fear of losing welfare benefits if they take a job, as well as improving their vocational and job-seeking skills, etc. The question arises as to how best tomaintain economies of scale and other efficiencies in each of the differentsupports given to a client while making them coordinated and flexibleenough to meet the needs of the clients realistically, in the right time, placeand manner. Is it best through merging the different services, or through better joint working (partnership) between different agencies (in which case theincentives must promote the flexibility required for effective joint working,while not introducing significant inefficiencies)?

While organization providing training and support for unemployed people under contract to the UK government Public Employment Service,Jobcentre Plus (under the policy during the 2000s termed “New Deal”, whichtargeted different groups of unemployed people), tend to be referred to as “partners”, the differential financial power, and control of resources and policy direction that characterize these providers’ relationships with thePublic Employment Service raise questions about models of partnership, andthe potential benefits and problems associated with different approaches(Lindsay and McQuaid 2008). Nevertheless, previous analyses of various models of partnership working and interagency cooperation generally pointto a number of benefits and limitations associated with such processes (seeMosley and Sol 2005). It is to these issues that we now turn.

Benefits of partnership and interagency cooperation

Partnership-based approaches to dealing with social and labor-market exclu-sion have become increasingly popular among policy-makers. A review ofthe literature suggests that there are a number of benefits associated withinteragency cooperation (Hutchinson and Campbell 1998; McQuaid 2000;Dowling et al. 2004; McQuaid et al. 2005).

Flexible and responsive policy solutions

Perhaps the most regularly deployed argument in favor of partnership-basedapproaches is that problems such as urban regeneration or labor-market exclusion are complex and multi-dimensional, requiring a range of inputs from stakeholders involved in delivering on social, economic and physicaldevelopment (Rhodes et al. 2003). The individual barriers (e.g. lack of skills),personal circumstances (e.g. caring responsibilities) and socio-economiccontext (e.g. living in an area of multiple deprivation and low job oppor-tunities) faced by people with low employability are often inter-related,overlapping and mutually reinforcing. Hence policy solutions aimed at onefactor, or part of the support system, are unlikely to be fully successful owing

to the counteracting impacts of other factors. Partnerships between key actorsor service providers are therefore essential in order to tackle the various causes as well as the symptoms of low employability. In terms of labor-market policies, local partnerships arguably facilitate the tailoring of the program and its delivery to the specific problems and opportunities of locallabor markets (Nativel et al. 2002).

Facilitating innovation and evaluation

Partnerships arguably have greater scope to test new and innovativeapproaches, as stakeholders coming together from a range of different policy perspectives can, in itself, produce greater dynamism through the sharing of ideas, expertise and practice, and risks can be contained. Theyalso allow individual partners to test new approaches and, if necessary, withdraw from unsuccessful or difficult experiences. Effective partnership working therefore challenges existing approaches by bringing to bear experience from other sectors and organizations, and developing new waysof working (Nelson and Zadek 2000). Under employability programs to assistunemployed people, policy-makers in the UK have sought to encourage the development of a flexible program that can operate slightly differentlyacross delivery areas, promoting experimentation and innovation, and theemergence of new ideas and solutions at the local level (DWP 2004). Beyondthe obvious benefits of such an approach, the flexible nature of localized partnerships facilitates a process of comparison and appraisal, so that bestpractice can be identified and alternative options and design features can beevaluated.

Sharing knowledge, expertise and resources

A defining feature of any interagency partnership is the manner in which skills, knowledge and expertise are shared in order to maximize the appro-priateness, quality and efficiency of provision. By engaging with private and third-sector providers with expertise in specific areas of service pro-vision, or with experience in engaging particularly disadvantaged clientgroups, public agencies can expand the reach, diversity and quality of theirservices.

Pooling of resources, synergy

At the most basic level, partnership-based approaches can increase the totallevel of resources brought to bear on problems, by increasing the numberof budget-holding organizations involved in delivering solutions (Conway 1999;McQuaid 1999). Synergy may also be achieved through combining comple-mentary resources from different organizations and from them operating in

more appropriate ways compared to their normal organizational approach.In addition, targeting or altering mainstream expenditure on specific sharedgoals (i.e. “mainstreaming” or “bending the spend”) and achieving synergies,and so maximizing the impact of resources, are issues that partnerships poten-tially have great impact.

Hence supporters of partnership-based approaches are particularly com-mitted to the idea that an effective partnership amounts to “more than asum of its parts”. Miller (1999), drawing on the example of effective localregeneration programs in England, enthuses about the potential for partner-ship to provide: “added value through the synergy of joint working” and a“transformational” learning process where stakeholders learn from eachother and often alter their own approach accordingly. “Effective partnershipscan be expected to generate: information sharing; improved communication;a better understanding of what each stakeholder can offer; the avoidance of duplication and inefficiencies; and the identification of opportunities foreffective sharing of resources” (Miller 1999: 349).

For Nelson and Zadek (2000), the achievement of this synergy or “partner-ship alchemy” depends upon five key factors: context, the socio-cultural environment and key drivers (systemic and specific triggers) that shape thecreation of partnership; purpose, the complexity and scope of partnershipgoals and activities, including the level of agreement on a common agenda;participants, the leadership characteristics, resources, capacities and com-petencies of different participants; organization, the organizational and legalstructure, governance principles and communication, consultation and conflictresolution mechanisms; outcomes, the ability to identify and evaluate out-comes and adapt the partnership accordingly.

Finn (2000) argues that national-government initiatives have often beenstructured in order to access the experience of working in the communityand engaging certain client groups held by local authorities and commu-nity stakeholders (including the former’s experience as large employers in many areas). By engaging with private and third-sector providers withexpertise in specific areas of service provision (e.g. mentoring or literacy training) or with experience in engaging particularly disadvantaged clientgroups, public agencies can expand the reach, diversity and quality of theirservices.

Developing a coherent service

Partnership working at the strategic level can ensure that policy initiativesin major areas of government activity are “aligned”. The drive to achievecoherent local and regional frameworks linking regeneration and employmentpolicies has been a major theme of recent reforms in the UK. The integra-tion of policies under shared strategic priorities can ensure that – for example– supply-side and demand-side labor-market strategies try to compliment each

other, and that supply-side interventions are informed by an understandingof the long-term needs of communities and local labor markets. There area number of examples of national employability policies being linked to local demand-led strategies and regional regeneration initiatives (Lindsay andSturgeon 2003; Gore 2004), but progress in this area has been uneven.

Improving efficiency and accountability

One of the key benefits associated with effective interagency cooperation is that it can lead to more efficient policy delivery, by eliminating the duplication of effort and improving communications. Within partnerships,interagency bodies have the capacity to be more democratic – at best theycan open up decision-making processes and gain the input and “buy-in” of organizations representing a broad range of constituencies and interests.However, there are often concerns that it is not clear “who is in charge” (see below).

Capacity-building

Examples of best practice in regeneration projects in England have demon-strated that local partnerships can build community capacity and engendera sense of community ownership (Rhodes et al. 2003). For the voluntary sector, interagency cooperation (particularly with government) offers newopportunities to have a practical impact on the policy agenda, enabling organizations to fulfill the key objectives of representing the community and giving voice to the concerns of disadvantaged groups (Miller 1999).Becoming “delivery partners” has also helped these organizations to accesslong-term and stable funding. At the strategic level, cooperation between government agencies and departments can lead to a sharing of knowledgeand practice across different areas of expertise, and result in joint workingtoward a seamless multi-faceted service for clients.

Gaining legitimization and “buy-in”

The tapping of “local knowledge” through the involvement of community-level stakeholders can contribute to the development of approaches that areable to engage disadvantaged communities and address specific localized problems. Engaging community-level stakeholders can also result in the legitimization of, and mobilization of local support for, new policy goals.This can be particularly important in geographical areas characterized bysevere disadvantage, where public agencies may be viewed with skepticismby residents who have previous experience of unsuccessful employability and regeneration interventions. By using local people to help in the develop-ment and implementation of policies, there can be greater ability to get the

target client group to help the recruitment and retention of “hard to reach”individuals, etc. (McQuaid 1999). At the planning level, where lead agenciesare willing to cede and share decision-making, budgets and responsibilities withpartners, they can engender a sense of shared ownership, helping to legitimizetheir policy aims.

The benefits discussed above may be achievable where effective structuresfor interagency cooperation and/or partnership working are in place. How-ever, there are considerable challenges in achieving these positive outcomeswhich are now discussed.

Partnership and interagency cooperation: potential problems and limitations

Partnerships may not achieve the potential for synergy owing to inertia or other reasons (e.g. Huxham 2003). Some of the challenges in achievingeffective and efficient partnerships are now discussed: a lack of clear and/orconsistent goals; resource costs; impacts on other services; and differencesin approach between partners. The specific issue of community participationis then discussed.

Conflict over goals and objectives

A lack of clear, specific aims or goals is often cited as a major cause of thefailure of partnerships. Many partnerships have agreed broad aims, but theirdetailed goals may be unclear or the partners may have differing understandingsof what the goals mean (Mitchell and McQuaid 2001). This can rapidly lead to misunderstanding, lack of coordination, and possible conflict between the partners. This may be accentuated if some partners have undeclared or “hidden” agendas. At the strategic level, conflicting priorities and “turfwars”, where different agencies fight over control of an issue or service, can undermine attempts at developing collaborative approaches. At theoperational level, gaining the commitment and engagement of private andthird-sector partners, and community representatives, can be complicated byorganizational barriers and inflexibilities, and localized problems in relationto limited community capacity.

Resources costs

There are considerable resources costs, for instance in terms of staff time in meetings and discussions and making agreements, and in delays to deci-sions owing to consultation with partners. It may also be difficult to closean inefficient or unsuccessful partnership, or even one whose objective has been achieved, if all partners do not agree, as this may “sour” relationselsewhere.

Accountability

There can also be problems of accountability as no single partner feels fullyaccountable for the actions of the partnership owing to the split betweenresponsibility and control (e.g. no single body takes full responsibility forproblems or for ensuring that overall the policy is effective and efficient). It may not be clear “who is in charge”. If each partner “claims” the full success of the partnership (e.g. in an initiative to help unemployed peopleseeking work) but only considers its own costs, this may distort decisions,and efficiency and value for money will be difficult to measure. The oppor-tunity or direct costs of staff time in participating in the partnership alsoneeds to be accounted for. The full social costs of the partnership need tobe aggregated and compared with the full social benefits, rather than eachpartner focusing upon its own costs and benefits (e.g. possibly through a formof social return on investment).

Impacts upon other services

Partnerships (especially those with stand-alone implementation units) maybe seen as an alternative to re-aligning mainstream services to deal with particular issues. But the scale of, and integration between, mainstream services may have a far more significant impact, especially in the long term.Conversely, partnerships may draw resources from other mainstream servicesor confuse the services in the minds of users, so reducing their effectiveness(i.e. there may be a significant opportunity cost in economic terms).

Organizational difficulties

Organizational difficulties inhibiting successful coordination of programs andapproaches, and overcoming the specialist concerns of disparate organizations,is a key implementation problem faced by agencies working together. Withinthis context, barriers to effective partnership working include: organizational(these include differing missions, professional orientations, structures and processes of agencies); legal/technical (statutes or regulations set down byhigher authority, and the technological capacity and practice of the organ-ization); and political (the external political environment but also internalbureaucratic politics).

At the strategic level, effective interagency cooperation may be underminedby the rigidity of institutional and policy structures. Government departmentsand agencies have arguably traditionally operated in narrowly focused “policysilos” (see above), with resources and expertise concentrated in specializedareas of policy. Breaking out of these policy “silos”, to develop multi-policyinteragency solutions, can be difficult owing to institutional arrangementsthat regulate the use of funding and deployment of manpower resources. There

is also a danger that strategic-level partnerships can be drawn into the minutiae of process, rather than focusing on implementing change – the actualoutcomes achieved by partnership working can be difficult to measure,which can lead to the partnership being seen as an end in itself rather thana means of implementing policy change (Ball and Maginn 2005).

Capacity-building and gaps

There can be difficulties when government seeks to engage different sectorsin delivering policy, if key stakeholders lack the professional, organizationalor financial capacity to contribute. There have been problems where govern-ments have sought to outsource provision before sufficient private- or voluntary-sector capacity is available. In many localities a lack of “com-munity capacity” (i.e. of the local people in the community) consistently undermines the ability of local stakeholders to engage in partnerships.Where local partnership structures are weak, a considerable commitment oftime, effort and resources is likely to be required in order to build capacity.As noted above, even with such a commitment, building trust may provedifficult in disadvantaged communities where public-service providers can be viewed with suspicion. Preparation of local communities (and others, suchas local employers) to participate effectively in partnerships often needs a clear strategy that is adequately resourced (and includes practical aspectssuch as being prepared in advance to deliver quick “wins” without waitingfor the usual long timescale of public-sector decision-making). Even at a professional level there is often a capacity gap in terms of specific skills andattitudes that hinders partnership development and implementation.

Differences in philosophy among partners

There may be significant differences in philosophy between the partners, suchas in the degree to which they feel the market can solve problems around a particular policy (e.g. employability) or the legitimate role of different stakeholders. There are a variety of related factors that have affected the development and implementation of partnerships, such as differing value and ethical systems between the public- and private-sector actors (OECD 2008;McQuaid 2000). Problems may arise in combining public and private management practices, philosophies and language within one partnership organization, while the extent to which formal contracting is a sound basisfor partnership has been debated. Contractualism offers benefits associatedwith accountability and clarity in responsibilities and reward structures. How-ever, where stakeholders are required to be both actors within a purchaser–provider contract and strategic partners there may be a confusion of rolesand incentives. It has also been suggested that the strict obligations asso-ciated with contractual relations (and even Service Level Agreements in the

public sector) can stifle some innovation. More generally, an integrated “policyculture” shared by agencies and groups involved in delivery is important ifpartnerships are to be effective. Where policy culture becomes fragmented– for example, owing to conflicting priorities over financial resources or tensions over the differential power of partners to “drive the agenda” – partnership working can quickly disintegrate (Dobbs and Moore 2002).

Power relations

The handling of differences in the relative power of different bodies or individuals in a partnership is important to its success. In most partnershipsthere is a degree of unequal power. The presence of unequal power shouldnot imply that all partners should necessarily have equal power. Some mayhave greater legitimate claim, due for instance to their greater involve-ment in the project or local area, or have greater political legitimacy in thecase of elected bodies. Although there are different types of power, greatestpower generally rests with those controlling resources. In the case of localregeneration partnerships, they are likely to dominate those in the local areawho may have a considerable understanding of what is relevant and effec-tive, albeit from a local rather than macro-perspective, and whose feeling of “ownership” can be crucial to the initiative’s success. At different stagesof a partnership there will be different balances of power between actors.To illustrate, in the early stages when an initiative is being developed, all those “around the table” will have potentially large influence as their involvement will often be considered important for getting the initiative started. However, the environment within which the key funders operate isvery influential (for instance, in ruling certain approaches out of discus-sion). When the initiative is agreed, then the views of the main funders arelikely to become relatively more important, i.e. there may be a shift fromthe influential power of some actors (such as local voluntary groups). AsMcDonald (2005) argues, power relations remains an area for greater theoretical development.

Community participation

In regeneration partnerships, a lack of “community capacity” consistentlyundermines the ability of local stakeholders to engage in partnerships(Dobbs and Moore 2002). Where local partnership structures are weak, aconsiderable commitment of time, effort and resources is likely to be requiredin order to build capacity (Rhodes et al. 2003). Carley (2006) notes the import-ance of an integrated “policy culture” shared by agencies and groupsinvolved in the delivery of local inclusion strategies.

The inclusion of community and voluntary-sector stakeholders can alsoraise issues surrounding the changing role of such organizations. McLaughlin

and Osborne (2000) argue that the increasing involvement of the voluntarysector as a delivery partner “risks subverting the legitimate role” of com-munity organizations by attaching them to “labyrinthine” contractual pro-cesses of regeneration programs. Osborne (1998) has also warned against thedanger of community-level actors becoming the “puppets” of governmentagencies, which may be attracted to indulging in tokenistic forms of localconsultation rather than less comfortable discussions with street-levelgroups representing the full diversity of community interests. The concernhere is that local authorities and other governmental stakeholders tend to be reluctant to cede authority, and seek “safe” forms of local engage-ment in which their decision-making autonomy is not challenged (Rowe andDevanney 2003).

Miller (1999) discusses the problems of government–community-sector cooperation from the opposite perspective, noting that the introduction of new stakeholders presents new potential problems beyond the familiar tensions between the state and private sectors. Miller distinguishes betweenthe formalized professional nonprofit sector and the “community sector” oflocal, informal and citizen-based organizations. He further distinguishesbetween primarily service-oriented organizations and those seeking to“enhance local democracy” (i.e. those with a political agenda). The formerprioritize the extent to which the partnership enables the delivery of services;the latter will be more interested in strategic action and the partnership process in itself.

It is crucial that advocacy groups engaged in practical partnerships arepersuaded to commit to constructive debate within the boundaries of the relevant policy agenda, and to consider their contribution to the delivery ofoutcomes. Without such a commitment, community-level stakeholders riskacting as a brake on progress, “putting themselves between regeneration agencies and local people” (Sanderson 1999). There can also be questions of legitimacy in the engagement of community-level stakeholders – govern-ment agencies seeking to build partnerships are understandably attracted to established community organizations, but these “usual suspects” may not always reflect the diversity of interests within local areas (Geddes 1998).

Some key success factors in partnership working

The above discussion highlights some of the advantages, problems andissues around partnership working and other approaches to interagency cooperation. Much of the literature on interagency cooperation emphasizesquestions of partnership structure, strategy and internal regulations (e.g.Hudson and Hardy 2002). Although this provides a useful overview to the question of partnership, it is perhaps lacking in specific examples of howsuccessful partnerships have emerged. Coupar and Stevens (1998: 145) state

that partnership “is not so much about institutions or methods, as aboutattitudes and culture. It is a question of building mutual trust, of recognis-ing differences and finding common ground”. This section identifies lessonsfrom successful partnerships and interagency initiatives, in which a numberof recurring features can be identified.

A clear strategic focus

Successful models of interagency cooperation tend to be governed by: adetailed, clearly defined strategy; a commitment to shared objectives and clear targets informed by an overarching strategic vision; a transparency of operations; and strategic interests being given priority over local or sec-tional interests. Rhodes et al. (2003), reviewing Single Regeneration Budgetpartnerships, emphasize the need for “formal sign-up” to an agreed strategyand approach from all relevant partners. Establishing “shared values” hasbeen a positive characteristic of local employability partnerships (Blaxter et al. 2003), although Miller (1999) suggests that this is more likely to succeed if cooperation is based around clearly defined, specific and limitedobjectives.

Strategic leadership and support

It is essential that there is clear strategic leadership and support for part-nership within each partner organization. Staff on the ground must beconfident of such support and be able to “speak for the organization” at mainpartnership meetings. This requires confidence in, and support for, staff fromsenior management and decision-makers to allow staff to make the partner-ship work effectively and efficiently. There must be a genuine willingness to make the partnership work, which may help to counteract the commontendencies to retreat into “policy silos” based on professional discipline ororganizational structure.

The importance of trust, organizations and people in partnerships

The importance of trust (between organizations and between individuals in partnerships) is often highlighted (Gambetta 1988). Effective delivery partnerships need: the right mix of skills and expertise; certainty within each partner organization regarding roles and responsibilities; continuity of approaches and membership in order to maintain “trust and certainty”;and a recognized and legitimate role for all partners, with no one actor dominating. Referring to regeneration partnerships, Rhodes et al. (2003) suggest that all immediately relevant policy actors must either directly signup to the specific project or be linked through wider partnership bodies for interagency cooperation to be effective. However, it is important for

partnerships to be clearly focused – Blaxter et al. (2003) and McQuaid et al. (2005) reflect on innovative local employability projects, noting thatonly appropriate stakeholders with the power, skills or resources (includingnetworks of influence) to add value to the partnership were included.

Capacity for cooperation and mutualism

Effective partnership or interagency cooperation operates through strong and established networks of communication and joint working at the locallevel, and – where external actors are involved – the inclusion of organ-izations with the capacity and resources to engage effectively in partnershiprelations and add value to the partnership process. It is essential that organizations, and individual representatives, involved in partnerships have both the authority and the institutional flexibility to engage in mutualdecision-making and resource-sharing. Training staff to participate effec-tively and efficiently in partnerships is essential for those involved in eitherthe development or the implementation of partnerships. Specific practical training should be provided to all staff involved (preferably jointly involv-ing staff from the relevant partners so that they can develop a common vocabulary and understanding and agreements on how to operate). Giventhe highly structured institutional framework of many public agencies, this can be problematic, but successful local employability partnershipshave seen actors such as the Public Employment Service work creatively to ensure maximum flexibility in the sharing of information and resources(McQuaid et al. 2005).

Organizational complementarity, co-location and coterminosity

The engagement of organizations that compliment each other’s resources andexpertise is important to maximizing the benefits of partnership working.Employability is a multi-dimensional issue, affected by individual factors, personal and family circumstances, and external barriers (McQuaid andLindsay 2005). It is therefore essential to ensure that there is a good “match”between the organizations represented in partnerships, so that a range of issuesaffecting the employability of different individuals and communities can beaddressed. At a practical level, there are benefits associated with the indi-viduals represented within partnerships holding similar levels of budgetaryand policy responsibility, and (where possible) operating within coterminous– or at least similar – geographical boundaries. Halliday and Asthana(2005), drawing on the example of Health Action Zones in rural areas of theUK, note how a lack of coterminosity and problems of physical distance can combine to constrain the development of the organizational capacity necessary to support community-based change. Co-location for the deliveryof services may also be beneficial in many cases.

Incentives for partners and “symbiotic interdependency”

If partnerships are to be effective, then mutual benefit and reciprocity areusually important. Actors must believe that there are benefits for their ownorganization set against the costs of involvement (benefits could include financial leverage, expansion of competencies and influence, achievement of organizational goals, positive public relations, or the opening of new markets). The presence of common or complementary goals is important,as is the degree of symbiotic interdependency – the extent to which benefitsfor one partner agency produce mutually beneficial outcomes for other part-ners (Fenger and Kok 2001). This contrasts with competitive interdependency,where the action of one actor interferes with another actor’s ability to takeaction or achieve his goals, potentially generating conflict. Systematically building and strengthening the “shadow of the future”, whereby future relationships are important and so influence current reciprocity, can reinforcepartnership working and the sharing of current benefits.

For example, effective employability partnerships tend to be built uponevidence of mutual benefits related to (for example) securing markets, and addressing recruitment and retention issues (for employers); and the extension of partnership working and areas of influence and competence, andthe achievement of positive outcomes for target groups (for policy actors andservice providers). Where the interdependency of mutual goals and benefitsis unclear, individual agencies may resist moves toward new cooperative structures.

The value of action and outcome-oriented procedures

Effective partnerships tend to focus on outcomes rather than merely on evidence of activity (Rhodes et al. 2003). In terms of the delivery of localemployability services, there is a need for: an emphasis on the quality as well as the quantity of outcomes; clear decision-making procedures, with management close to service provision; and an action- and results-orientedapproach, with measurable goals clearly defined and evaluated (McQuaid et al. 2005). However, we should finally acknowledge that measuring the value of effective interagency cooperation is likely to prove difficult. Thereare considerable problems in seeking to identify the specific impacts of anyone model of cooperation. Nickell and Van Ours (2000: 219), referring to Dutch and UK labor-market policies, note the problems associated withcomparing the impact of different “configurations of institutions”:

First, there is no empirical basis to disentangle the separate con-tribution of each policy change. Second, some policy changes aretime consuming and therefore time lags may be substantial. . . . Third,policy changes are complimentary. The effect of one policy depends

on whether or not a different policy is implemented as well. A changeof institutions in the labour market is a package deal.

In general, outcome-oriented partnerships are characterized by: an emphasison the quality as well as the quantity of outcomes; responsiveness and cleardecision-making procedures, with management close to service provision; anda consistent approach to reviewing results, with measurable goals clearly definedand evaluated at regular, appropriate intervals.

“New Governance” and “New Public Management”

Geddes and Benington (2001), reviewing social-inclusion strategies across the EU, suggest that specific institutional configurations are particularly associated with certain types of welfare state. Statist, interventionist welfarestates, such as those found in Nordic states, are less likely to develop broadmulti-agency partnerships with nonstate organizations, and more likely to deploy government agencies as the main or sole provider of activationemployment policies. However, shifts toward new “state–market–civil society” mixes in the provision of welfare and employability services are common to a range of different welfare state models (Van Berkel and Vander Aa 2005; Lindsay and McQuaid 2008).

These shifts to some extent reflect a more general move toward new formsof public-sector governance. Governance can be defined as the frameworkthrough which political, economic, social and administrative authority is exercised at local, national and international levels. This framework consistsof a wide variety of mechanisms, processes, institutions and relationships(including partnerships) through which individual citizens, groups and organ-izations can express their interests, exercise their rights and responsibilities,and mediate their differences (Nelson and Zadek 2000). Governance isincreasingly about balancing the roles, responsibilities, accountabilities andcapabilities of: different levels of government – local, national, regional andglobal; and different actors or sectors in society – public, private and civilsociety organizations and individual citizens.

Moves toward what has been termed the “new governance” – charac-terized by a shift in the roles and responsibilities bureaucracies, and the involvement of private agencies in service delivery – stem partly from concerns over budget constraints on the public sector, higher client expectations and therefore the demand for better-quality services, the drivefor efficiency through “least cost, best performance” approaches, and the belief that private-sector management systems can deliver these benefits(Considine 2000).

Rhodes (1997) argues that New Public Management (NPM) is one manifestation of these new forms of governance. NPM is characterized bythe deployment of business principles and management techniques, and the

use of private enterprise to deliver public services, partly to gain associatedefficiencies. This may be done through partnerships combining public- andprivate-sector bodies or through quasi-markets and purchaser–providersplits in the organization of public services. Exponents have seen NPM asoffering a solution through more flexible organization, flattened managementhierarchies, and the decentralization of decision-making (Hood 1991).

Such approaches emphasize the importance of choice for the provider of a public service and the implementation schemes to exploit possibleefficiency gains in the provision of public services. This, however, is a more restricted view of partnerships than has been taken in this chapter. In the case of Private Finance Initiatives (repackaged as Public–PrivatePartnerships in the UK after the new government of the late 1990s), theremay be no increase in choice of service or product for the ultimate user of the service. This partly reflects the outcomes of the debates since the 1980s concerning whether the public sector should have an enabling role,determining the form and level of public services but not primarily deliver-ing them, or a role as sole provider of services (see, for instance, Giloth andMier 1993).

There remain questions of accountability, with one argument being thatlocal authorities and government departments are more clearly and publiclyaccountable than multi-agency quangos and contracted providers. Others arguethat the introduction of quasi-markets, in an attempt to stimulate market-type competitive behavior, is problematic, as such behavior emerges only fromthe operation of genuinely open markets, but the product monopolies andsingle state purchasers of public services that characterize “internal market”arrangements do not provide such an environment (Drechsler 2005). The critique of NPM is that it attempts to replace poor public management with private-sector inputs rather than with better public management (Beviret al. 2003), and marks an attempt to reduce costs (with implications for services) by taking elements of public spending “off the books” (Newmanand McKee 2005).

With the expansion of this contractualism through competitive tenderingin employment and other policy areas, it has been suggested that the mannerin which service providers have been required to engage in such processesof compulsory competitive tendering has created new rivalries, which havethe potential to undermine the ethos of partnership (Peck 2001). The form,power structure and implementation of partnerships is key. In the case ofUK employment policy in the late 1990s and the 2000s, there is some evidence that the process of strategic delivery plan development early in theprogram’s rollout was undermined by the competitive arguments made byorganizations in favor of an expansion in their own role in the delivery pro-cess (Mason 2007). However, despite some early claims that the extensionof the private sector’s role into unfamiliar areas of employability provisionunder the UK’s policies to assist unemployed people initially has proved to

be problematic, there is no consistent evidence of differences in quality as aresult of private-sector inputs (Hasluck 2001), while Dunleavy et al. (2006)have argued that NPM is in decline. More generally, the future of wider non-contractual-based partnerships and intergovernmental working may bemore assured, for the reasons discussed previously.

Conclusions

The above discussion highlights the complexity and diversity of the issuessurrounding the development and implementation of partnership approachesand other forms of interagency cooperation. The chapter provides a frame-work for considering the advantages and problems of partnership working,particularly in the context of regeneration and employability policies. It also discussed “critical success factors” in terms of: strategic focus; the participation of key individuals and organizations; a shared capacity for cooperation and mutualism; incentives and symbiotic interdependency;organizational complimentarity and coterminosity; and outcome-oriented pro-cedures. A number of specific issues have emerged from the above analysis,which revolve around lessons we can learn concerning how partnerships can be improved (in terms of strategic direction, structure, operation, etc.)and when are they appropriate.

If we are better to understand, and theorize on, when and how to improvepartnership working, then further research is needed into several sets of ques-tions, relating to why have partnerships and what form they should take.First, there is a need to identify the balance between a body carrying out its activities largely alone (with potential benefits of clearer accountability,speed of action and reduced transactions costs associated with partnerships)or in a partnership with other agencies (with potential costs and benefits dis-cussed above) or somewhere on the multi-dimensional spectrum between these.Second, for what reasons and under what circumstances are different formsof partnership most appropriate, and what are the implications of different forms of partnership working? It is important to identify the different typesof partnership based upon such factors as motivations, benefits and costs.We need to consider how, and to what extent, interagency cooperation has facilitated the development of innovative and locally responsive policy solutions and what forms of partnership organization most facilitate this indifferent macro- and micro-circumstances. Third, what benefits and problemshave been associated with the implementation of New Public Managementtype approaches to specific areas such as employability and regeneration policies, and what are the implications of different models of private-sectorparticipation? To what extent do different models of partnership working,or contracting out, contribute to effective interagency cooperation and, crucially, outcomes, and what are the tensions between contractualism andstrategic partnership working?

Given the complexity of issues concerning partnerships, there is a need to:clarify our typologies of partnerships working; develop a greater understandingof the issues concerning the implementation of partnerships on the ground;and create more nuanced general theoretical analyses of partnerships.

Acknowledgments

Many of these ideas were developed through discussions with Colin Lindsayand other colleagues, including Matthew Dutton and Malcolm Greig, at theEmployment Research Institute, and this is gladly acknowledged.

References

Bailey, N., Barker, A. and MacDonald, K. (1995) Partnership Agencies in British UrbanPolicy, London: UCL Press.

Ball, M. and Maginn, P. J. (2005) “Urban Change and Conflict: Evaluating the Roleof Partnerships in Urban Regeneration in the UK”, Housing Studies, 2 (1): 9–28.

Bevir, M., Rhodes, R. A. W. and Weller, P. (2003) “Traditions of Governance: Inter-preting the Changing Role of the Public Sector”, Public Administration, 81 (1): 1–17.

Blaxter, L., Farnell, R. and Watts, J. (2003) “Difference, Ambiguity and thePotential for Learning: Local Communities Working in Partnership with LocalGovernment”, Community Development Journal, 38 (2): 130–9.

Carley, M. (2006) “Partnership and Statutory Local Governance in a DevolvedScotland”, International Journal of Public Sector Management, 19 (3): 250–60.

Commission of the European Communities (CEC) (2001) Strengthening the LocalDimension of the European Employment Strategy, Luxembourg: Office for OfficialPublications of the European Communities.

Commission of the European Communities (CEC) (2003) The Future of the Euro-pean Employment Strategy: A Strategy for Full Employment and Better Jobs forAll, Luxembourg: Office for Official Publications of the European Communities.

Considine, M. (2000) “Contract Regimes and Reflexive Governance: ComparingEmployment Service Reforms in the United Kingdom, the Netherlands, NewZealand and Australia”, Public Administration, 78 (3): 613–38.

Conway, M. (1999) Partnerships, Participation, Investment, Innovation: Meeting theChallenge of Distressed Urban Areas, Dublin: European Foundation.

Couper, W., Stevens, B. (1998) “Towards a New Model of Industrial Partnership”,in Sparrow, P., and Marchington, M. (eds), Human Resource Management: TheNew Agenda, London: Financial Times Pitman Publishing, pp. 145–59.

Department for Work and Pensions (DWP) (2004) Building on New Deal, Sheffield:DWP.

Department for Work and Pensions (DWP) (2006) A New Deal for Welfare: EmpoweringPeople to Work, London: DWP.

Dobbs, L. and Moore, C. (2002) “Engaging Communities in Area-based Regenera-tion: The Role of Participatory Evaluation”, Policy Studies, 23 (3–4): 151–71.

Dowling, B., Powell, M. and Glendinning, C. (2004) “Conceptualising SuccessfulPartnerships”, Health and Social Care in the Community, 12 (4): 309–17.

Drechsler, W. (2005) “The Rise and Demise of the New Public Management”, Post-autistic Economics Review, 33 at: www.paecon.net/PAEReview/

Dunleavy, P., Margetts, H., Bastow, S. and Tinkler, J. (2006) Digital Era Govern-ance: IT Corporations, the State and E-government, Oxford: Oxford University Press.

Fenger, M. and Kok, P. J. (2001) “Interdependency, Beliefs, and Coalition Behavior:A Contribution to the Advocacy Coalition Framewoork”, Poplicy Sciences, 34 (2):157–170.

Finn, D. (2000) “Welfare to Work: The Local Dimension”, Journal of European SocialPolicy, 10 (1): 43–57.

Gambetta, D. (1988) “Can We Trust Trust?”, in D. Gambetta (ed.) Trust: Makingand Breaking Cooperative Relations, Oxford: Blackwell.

Geddes, M. (1998) Local Partnership: A Successful Strategy for Social Cohesion?, Dublin:European Foundation.

Geddes, M. and Benington, J. (2001) “Social Exclusion and Partnership in the EU”,in M. Geddes and J. Benington (eds) Local Partnerships and Social Exclusion inthe European Union: New Forms of Local Social Governance?, London: Routledge.

Giloth, R. and Mier, R. (1993) “Co-operative Leadership for Community ProblemSolving”, in R. Mier et al. (eds) Social Justice and Local Development Policy, NewburyPark, Calif.: Sage.

Gore, T. (2004) “The Open Method of Co-ordination and Policy Mainstreaming: TheEuropean Employment Strategy and Regional Conversion Programmes in the UK”,European Planning Studies, 12 (1): 123–41.

Halliday, J. and Asthana, S. (2005) “Policy at the Margins: Developing CommunityCapacity in a Rural Health Action Zone”, Area, 37 (2): 180–8.

Harding, A. (1990) “Public–Private Partnerships in Urban Regeneration”, in M. Campbell (ed.) Local Economic Policy, London: Cassell.

Hasluck, C. (2001) “Lessons from the New Deal”, New Economy, 8 (4): 230–4.Hood, C. (1991) “A Public Administration for All?”, Public Administration, 69 (1):

3–17.Hudson, B. and Hardy, B. (2002) “What Is Successful Partnership and How Can It

Be Measured?”, in C. Glendinning, M. Powell and K. Rummery (eds) Partner-ships, New Labour and the Governance of Welfare, Bristol: Policy Press.

Hutchinson, J. and Campbell, M. (1998) Working in Partnership: Lessons from theLiterature, DfEE Research Report RR 63, London: Department for Education andEmployment.

Huxham, C. (2003) “Theorizing Collaborative Practice”, Public Management Review,5 (3): 401–23.

Knox, C. (2002) Review of Public Administration: Partnerships, Belfast: Office of theFirst Minister and Deputy First Minister.

Lankshear, C., with Gee, J. P., Knoebel, M. and Searle, C. (1997) ChangingLiteracies, Buckingham: Open University Press.

Lindsay, C., and McQuaid, R. W. (2008) “Inter-agency Co-operation in Activation:Comparing Experiences in Three Vanguard ‘Active’ Welfare States”, Social Policyand Society, 7 (3): 353–65.

Lindsay, C., McQuaid, R. W. and Dutton, M. (2008) “New Approaches toEmployability in the UK: Combining ‘Human Capital Development’ and ‘WorkFirst’ Strategies?”, Journal of Social Policy, 36 (4): 539–60.

Lindsay, C. and McQuaid, R. W. (2009) “New Governance and the Case of Activa-tion Policies: Comparing Experiences in Denmark and the Netherlands”, SocialPolicy and Administration, 43 (5): 445–63.

Lindsay, C. and Sturgeon, G. (2003) “Local Responses to Long-term Unemployment:Delivering Access to Employment in Edinburgh”, Local Economy, 18 (2): 159–73.

McDonald, I. (2005) “Theorising Partnerships: Governance, Communicative Actionand Sport Policy”, Journal of Social Policy, 34 (4): 579–600.

McLaughlin, K. and Osborne, S. P. (2000) “A One Way Street or Two Way Traffic?Can Public–Private Partnerships Impact on the Policy-making Process?” in S. P. Osborne (ed.) Managing Public–private Partnerships for Public Services: AnInternational Perspective, London: Routledge.

McQuaid, R. W. (1999) “The Role of Partnerships in Urban Economic Regenera-tion”, International Journal of Public–Private Partnerships, 2 (1): 3–28.

McQuaid, R. W. (2000) “The Theory of Partnerships – Why Have Partnerships?”,in S. P. Osborne (ed.) Managing Public-private Partnerships for Public Services:An International Perspective, London: Routledge.

McQuaid, R. W. and Lindsay, C. (2005) “The Concept of Employability”, UrbanStudies, 42 (2): 197–219.

McQuaid, R. W., Lindsay, C., Dutton, M. and McCracken, M. (2007) Best Practicein Inter-agency Co-operation on Employability, report for the Department forEmployment and Learning, Northern Ireland, Belfast: DELNI.

McQuaid, R. W., Lindsay, C. and Greig, M. (2005) “Job Guarantees, EmployabilityTraining and Partnerships in the Retail Sector”, Local Economy, 20 (1): 67–78.

McQuaid, R. W. and Scherrer, W. (2010) “Changing Reasons for Public PrivatePartnerships”, Public Money and Management, 30 (1): 27–34.

Mason, M. (2007) “Collaborative Partnerships for Urban Development: A Study ofthe Vancouver Agreement”, Environment and Planning A, 39 (10): 2366–82.

Miller, C. (1999) “Partners in Regeneration: Constructing a Local Regime for UrbanManagement?”, Policy and Politics, 27 (3): 343–58.

Mitchell, I. and McQuaid, R. W. (2001) “Developing Models of Partnership inEconomic Regeneration”, in L. Montanheiro and M. Spiering (eds) Public and PrivateSector Partnerships – the Enterprise Governance, Sheffield: Sheffield Hallam Uni-versity Press.

Mosley, H. and Sol, E. (2005) “Contractualism in Employment Services: A Socio-economic Perspective”, in E. Sol and M. Westerveld (eds) Contractualism inEmployment Services, The Hague: Kluwer, http://www.ilo.org/public/english/bureau/inst /download/dp15604.pdf

Nativel, C., Sunley, P. and Martin, R. (2002) “Localising Welfare-to-work?Territorial Flexibility and the New Deal for Young People”, Environment andPlanning C: Government and Policy, 20 (6): 911–32.

Nelson, J. and Zadek, S. (2000) Partnership Alchemy: New Social Partnerships in Europe,Copenhagen: Copenhagen Centre.

Newman, J. (2001) Modernising Governance: New Labour, Policy and Society,London: Sage.

Newman, J. and McKee, B. (2005) “Beyond the New Public Management? PublicServices and the Social Investment State”, Policy and Politics, 33 (4): 657–73.

Nickell S. and Van Ours J. (2000) “Why Has Unemployment in the Netherlands andthe United Kingdom Fallen So Much?”, Canadian Public Policy, 26 (1): 201–20.

Organisation for Economic Cooperation and Development (OECD) (1990) Partner-ships for Rural Development, Paris: OECD.

Organisation for Economic Cooperation and Development (OECD) (2008) Public–Private Partnerships: In Pursuit of Risk Sharing and Value for Money, Paris:OECD.

Osborne, S. (1998) “Partnerships in Local Economic Development: A Bridge TooFar for the Voluntary Sector?”, Local Economy, 12 (4): 290–5.

Osborne, S. P. (ed.) (2000) Managing Public–private Partnerships for Public Services:An International Perspective, London: Routledge.

Peck, J. (2001) Workfare States, New York: Guilford.Rhodes, J., Tyler, P. and Brennan, A. (2003) “New Developments in Area-based

Initiatives in England: The Experience of the SRB”, Urban Studies, 40 (8): 1399–1426.

Rhodes, R. A. W. (1997) Understanding Governance, Buckingham: Open UniversityPress.

Rowe, M. and Devanney, C. (2003) “Partnership and the Governance of Regener-ation”, Critical Social Policy, 23 (3): 375–97.

Sanderson, I. (1999) “Participation and Democratic Renewal”, Policy and Politics,27 (3): 324–41.

Snape, D. and Stewart, M. (1996) “Keeping up the Momenton – Partnership Work-ing in Bristol and the West of England”, Bristol: Bristol Chamber of Commerce.

Social Exclusion Unit (SEU) (2001) Preventing Social Exclusion, London: SEU.Van Berkel, R., and van der Aa, P. (2005) “The Marketisation of Activation Services:

A Modern Panacea? Some Lessons from the Dutch Experience”, Journal ofEuropean Social Policy, 15 (4): 329–345.

PUBLIC–PRIVATE PARTNERSHIPS AND PUBLICGOVERNANCE CHALLENGES

Carsten Greve and Graeme Hodge

Abstract

Public–private partnerships (PPPs) come in many different institutional forms,including those which use private finance and create long-term contracts forpublic infrastructure projects. This chapter reviews briefly the empiricalexperience with infrastructure PPPs in the OECD countries. The chapter thenproceeds to examine what kinds of challenges PPPs pose to the theory andpractice of public governance. In the discussion there is special emphasis on consequences of PPPs for democratic governance and for the prospectof genuine collaboration between the public sector and the private sector inthe future.

Introduction

How do public–private partnerships (PPPs) fit into the new public governanceregime? And what are the public governance challenges associated with therise of PPPs? PPPs can be loosely defined as institutionalized cooperativearrangements between public-sector actors and private-sector actors. Therehave been numerous attempts to classify, order and characterize PPPs in the literature (Hodge and Greve 2007). While a broad conception of PPPsmight be appropriate in some circumstances, this chapter examines a specifickind of PPP that occurred from the mid-1990s: long-term infrastructure contractual-type PPPs or, as the European Union terms them, “institution-alized PPPs”. PPPs as long-term infrastructure projects are one prominentunderstanding of PPPs that will be used here. These types of PPPs were introduced in the United Kingdom from the 1990s onwards (although thereare historical examples: see Wettenhall 2003, 2005; Kettl 1993), and soon caughtthe attention of various governments around the world, notably Australia.The PPP institution has been a small but important part of government reform

in most OECD countries as well as in developing countries (for overviews,see Osborne 2001b; Deloitte 2006; Hodge and Greve 2005; Ghobadian et al.2004; OECD 2008).

This chapter discusses how PPPs became a part of the new public govern-ance regime as they allowed governments to engage with a number of pri-vate agents in often complex and contractually sophisticated relationships.Governments have been using PPPs for various activities in order to tacklecross-border challenges through private partners. In a democratic context,the role of PPPs remains debatable. PPPs mean that governments may bein danger of detaching themselves from the wider electorate and democraticconstituencies at the grassroot levels because the PPP deals appear so com-plex and inaccessible that PPPs have become an elitist game – a game playedby economists, lawyers, consultants and financial experts.

The chapter is structured in the following way. The first part provides a brief overview of the recent PPP evolution and debate, and traces the development of PPPs from their primarily UK origin to the global attentionthey are enjoying currently. The second part discusses the challenges thatPPPs present for new public governance structures. The third part discussesif there is a democratic gap between the PPP deals and the democratic institutions. The fourth part concludes the chapter by assessing the role ofPPPs in future public governance regimes.

Public–private partnerships for infrastructure

PPPs in their modern-day incarnation as long-term infrastructure contractsare now a commonplace factor in political, administrative and economic reformdiscussions throughout the world. In a PPP, the government interacts with anumber of private-sector companies to design, finance, build, own and operateinfrastructure projects, and maybe transfer them back to the government again. There is a great variety in how these elements can be used, for exampleDesign Finance Build Operate (DFBO) or Build Own Operate Transfer(BOOT). The contractual periods are often long – up to thirty or forty years,or even longer. There can be considerable risk involved. But there is alsopossibility for innovation and gain-sharing. The key argument has been thatthe government and the private sector work together toward a common purposeful objective. One definition sees PPPs as “cooperation of some sortof durability between public and private actors in which they jointly developproducts and services and share risks, costs and resources which are con-nected with these products” (Van Ham and Koppenjan 2001: 598). Therehas been a discussion whether PPPs represent a genuine modernizationeffort, or if PPPs are just a mere excuse on the road to more privatization.The US scholar Stephen Linder’s (1999) early survey of the main argumentsrelated to PPPs mentioned the possibility that “PPP” was just another

word for privatization. While that discussion is not completely dead, PPPshave emerged in their own right in the public governance literature as onerecognizable organizational and institutional form and something morethan just marketization (Bovaird 2006).

The sectors in the UK where most PPPs have been located include health,defence, education and transport (HM Treasury 2003). The UK emerged as the primary exponent for the PPP phenomenon (Osborne 2001b). PPPswere a part of the John Major government’s reform program in the UK in the early 1990s. It started life as the Private Finance Initiative (PFI) at the time. There were several objectives with the Major government’s policy for private involvement in public infrastructure (Terry 1996). One objective was to reduce the need for public-sector borrowing by letting private finance come in. Another objective was to shift the balance betweenthe public sector and the private sector in favor of the latter sector. The PFI scheme was clearly a part of the Conservative government’s agenda for the public sector in the 1990s. In 1997, however, the New Labour government was elected in the UK. One option for the new government could have been to abolish the PFI scheme at the time. Instead, the NewLabour government chose to keep the PFI scheme. Why? Two reasons have been mentioned in the literature (Flinders 2005). First, New Labourneeded the private finance to help finance the welfare services that New Labour had promised as part of its broader modernization agenda. As New Labour had committed itself not to raise taxes, they needed a way to finance the modernization agenda. Second, New Labour could exploit thePPP theme to signal to the private sector that the party was not anti-business and nothing like the old Labour party. Whatever the cause, the PPPagenda received a boost with the election of the New Labour governmentin the UK.

There has been a steady rise in the number of projects. In 2008, there were 625 PPPs in the UK with a total capital value of £58.87 billion (HMTreasury 2008: 6). The Treasury has been the key driver in the UK PPP policy. The policy has been laid out in several key documents on the PFIscheme (HM Treasury 2003, 2006). The UK National Audit Office has followed the policy closely, and evaluated many of the projects as well asthe policy (National Audit Office 2003). The evaluations made by the NationalAudit Office are probably some of the most thorough evaluations one is likelyto find about PPPs in the world.

The debate in the UK has ranged from the early assessment of theInstitute for Public Policy Research (2001) report (which to a certain extentendorsed PPPs but also warned against the notion that PPPs were the onlyshow in town) to the present controversy over the financial implications and current high-profile projects (Pollock et al. 2007). For Flinders (2005),PPPs changed with the different political priorities of the New Labour

PUBLIC–PRIVATE PARTNERSHIPS

government during the 2000s. The arguments changed from being about economics to being about innovation. For Broadbent and Laughlin (1999),PPPs were a part of the ongoing UK modernization process in the publicsector. The academic assessment of whether PPP has been good value is atheme where there are some disagreements. Some reviews of the UK experi-ence have cast PPPs in a favorable light (Pollitt 2005). Others have remainedmore skeptical (Shaoul 2005). Recently, researchers have begun to questionthe validity of the presumed savings about which the UK Treasury has beenboasting (Pollock et al. 2002, 2007).

Australia was another place where PPPs caught on from the late 1990s.Many projects have been initiated in Australia, ranging from a tunnel in Sydney harbor to a motorway and tunnels in Victoria – the Citylink pro-ject. The Australian PPP experience has generally lived up to many of theofficial policy expectations (Allen 2007; Hodge 2005). But there have alsobeen failures along the way – in the hospital sector, for example (English2005).

In the rest of Europe, such as Germany, Spain and Italy, PPPs have alsobeen prolific. In Germany new hospitals and roads are being built as PPPs.In Spain, new transport facilities are being constructed as PPPs. In the newEuropean member states, PPPs have become an option to be considered(Bovaird 2004; Teisman and Klijn 2001). In the US and Canada, PPPs havebeen an important part of the strategy to renew infrastructure (Bloomfieldet al. 1998; Rosenau 2000) (see the recent OECD 2008 overview of what countries are doing in relation to PPPs).

In international reviews from advocates in consultancy firms (see, for example, Deloitte 2006; PriceWaterhouseCoopers 2005), PPPs are beingpraised. International organizations such as the European Union and theOECD are making cautious endorsements of PPPs. The international organ-izations admit that PPPs are viable tools to use, but they refrain from beingoverly enthusiastic. The OECD states that PPPs are one policy option out ofmany. Perhaps the international organizations have learned the lessons ofthe 1980s and 1990s where privatization was being lauded, with subsequentproblems occurring in the aftermath.

Summing up, the PPP has become an important new institution in the worldof infrastructure policy. While some countries, notably the UK and Australia,have been at the forefront of developments, other countries and internationalorganizations have followed suit more reluctantly. Many countries and alsointernational organizations are being cautious about the PPPs’ long-term perspectives, and are careful not to become too enthusiastic. Meanwhile, the bulk of scholars have been skeptical about the economic promises aboutPPPs, although some of the academic reviews to date have acknowledged thespecific economic and financial results obtained by using PPPs in particularcountries and sectors.

Public–private partnerships and public governance challenges

In many ways, PPPs contribute to the new style of governing emphasized inthe New Public Governance. As we have learned in the introduction to thisbook, the new public governance puts emphasis on different factors fromthose emphasized by New Public Management. In a governance framework,authority is spread among a number of actors (Heinrich and Lynn 2001).No one is quite in charge, and the governance responsibilities are dispersed(Kettl 1993). Governments have used PPPs to build relationships with theprivate sector and to reach public goals through private means. PPPs grewout of the privatization era, but PPPs could also be seen as a genuine new way of bonding between public-sector and private-sector organizations that would allow for more sophisticated use of private-sector expertise.Governments can tap into private-sector knowledge in order to innovate, sharerisks with private-sector organizations, and realize gains and share them with private-sector organizations in a way that was not possible in the privat-ization era. And they can make use of private-sector organizations when public-sector organizations are not capable of handling the tasks them-selves. Whether new hospitals or new primary schools were being built, ornew bridges or tunnels constructed, governments could use the private-sector expertise in innovative hospital-building, tunnel construction andschool-building to get higher-quality projects, the argument went. One keyfactor in aligning private-sector means with public-sector goals was to tiethe success of the projects closely to private finance. Private-sector investorswould see public infrastructure projects as safe and reliable investments. Public-sector organ-izations could use the private finance made available to them,and also use it to make sure that private-sector firms kept an interest in theprojects.

There are, however, many challenges to public governance that occur becauseof PPPs, and here we shall restrict ourselves to five of them.

The first challenge in governing through PPPs is complexity. PPPs havecome to be seen as increasingly complex deals. PPPs are long-term contracts,so they are shielded by sudden political interference. They are as suchmainly accessible to experts in finance and law, and can be very difficult forlay people to understand. Often the deals and contracts are not documentsthat engage publics in debates over the future governance challenges. PPPdeals rely on consultants to formulate and to check the deals. There are complex negotiations leading up to a PPP contract, and there is very seldom a process of inviting the public or other stakeholders to have their say in the process. More often than not, the emphasis has been to seclude the negotiations and keep them for the partners to the contract themselves. Thereis the related criticism that PPPs are often not transparent. The deals are

kept as secret as commercial contracts. They are rarely open for outsidescrutiny. The institutional variety of PPPs adds to the complexity of the deals themselves. A special organizational unit that embodies the partner-ships is often established as PPPs are set up, and in the UK this is knownas the Special Purpose Vehicle. Often the Special Purpose Vehicle consistsnot only of two distinct organizations, but also of a number of organiza-tions on each side of the negotiation table. A public-sector part may consistof several ministries and agencies working together on a particular project.A private-sector part may consist of a lead agency with many other organ-izations under contract to the lead agency. Add to that the financial institu-tions (banks and finance investors), the lawyers who oversee the contracts,and the various consultants who are drafted in to solve various technical issues, and there is a host of organizations involved in the partnership (Hodge2004a,b).

A second challenge is the number of roles governments adopt in PPP relationships. A key organization is often the Treasury/ministry of financesas the organization in charge of the process. But, as with privatization, there is a certain double act appearing here. A ministry of finance is bothan advocate of the PPP policy and a key guardian of the public purse. Oftenthese two roles would be expected to collide. The organizations that shouldsound cautions are also often highly involved in policy development of somesort, such as the national audit offices that are supposed to check the govern-ment’s policy, but also are involved in evaluations and recommendations togovernments on PPP policy. Hodge (2005) notes how governments have movedfrom their traditional stewardship role to a louder policy-advocacy role concerning PPPs. As a consequence of this, we might reflect that governmentnow finds itself in the middle of multiple conflicts of interest, acting in theroles of policy advocate, economic developer, steward for public funds, electedrepresentative for decision-making, regulator over the contract life, commercialsignatory to the contract, and planner.

A third challenge is when to choose a PPP over an alternative governanceform. A PPP is one option as a governance form, but there are others. Otheralternatives are hierarchy through a traditional public-service organization,contracting out/outsourcing, and finally a more or less loosely structured network. Choosing among the different governance forms is not always a deliberate action by governments. How to weigh up the advantages and disadvantages is often given too little thought. The closest alternative to aPPP arrangement would be contracting out/outsourcing. Some researchersbelieve that there is really no difference between the two, and that most ofwhat passes for PPPs actually is contracting out (Klijn and Teisman 2005).Whether governance forms could be matched to the governance forms in theway that Williamson (1985) suggests, or as Karmarck (2007) in her analysisof different governance forms for public policy challenges might envisage,is a topic of some controversy; but there may be a need for discussing PPPs

in this light instead of advocating PPPs for every policy area (see also theBritish Institute of Public Policy Research in 2001).

A fourth challenge for the government is how to manage partners fromthe private sector that follow their own strategic agenda. Many governmentsdo not possess enough capacity to govern the new PPPs; or, put in otherterms, the private sector as agents acts strategically in order to avoid the scrutinizing moves by the public-sector principals. Private-sector companiesmay try to outsmart governments, and other governmental partners, and notbe able to fulfill the contracts and the obligations into which they have entered.This kind of analysis would suggest a closer look at the market conditionsfor private-sector organizations entering into PPP arrangements. Private-sector organizations may be interested in PPPs in several ways, one of thembeing that they seek to exploit weaknesses in the public sector, and try toget lucrative deals on behalf of their owners and stockholders. Governmentsshould be able to oversee the ecology of a market (Frumkin 2002); but, ifthey cannot do so, private-sector agents may find ways to bypass governmentintentions and rules.

A fifth challenge is that a PPP is a long-term contract. A PPP is a particu-larly tricky institutional form because its long-term contracts (up to thirtyor forty years) make it almost by definition impossible to foresee which factors can influence the governance environment in the long run. More attention in the literature on public governance is being paid to developmentsover time (Pierson 2004; Pollitt 2007). PPPs are good examples of that.Although over 600 PPP projects in the UK, for example, have been estab-lished, it is difficult to evaluate the output and the outcome of these projectsbecause of the duration of the contractual period.

Added to this challenge has been the first peer review of the impressiveon-time and on-budget figures reported by Mott Macdonald (2002). The reviewof Pollock et al. (2007) was unequivocal in its judgment of these figures, stating:

there is no evidence to support the Treasury cost and time overrunclaims of improved efficiency in PFI . . . [estimates being quoted are]not evidence based but biased to favor PFI . . . only one study compares PFI procurement performance, and all claims based on[this] are misleading.

Boardman, Poschmann and Vining (2005), for instance, noted the diffi-culty of capturing transaction costs in any comparison between partnershipand traditional project delivery, and catalogued seventy-six major NorthAmerican PPP projects. They noted that less than half included a significantprivate financing role. They presented five transport, water provision and wasteprojects, showcasing a series of “imperfect” partnership projects with highcomplexity, high asset specificity, a lack of public-sector contract-management

skills and a tendency for governments to be unwilling to “pull the pin” onprojects once under way. They particularly point to private entities being “adept at making sure, one way or another, that they are fully compensatedfor risk-taking” and to strategic behavior such as declaring bankruptcy (or threatening to) in order to avoid large losses. There are clear tensionsfor governments here, having to hold their nerve and watch commercial failures materialize when risks are borne by the private sector, despite theiryearning to be viewed as successfully governing a growing and vibrant market. PPPs encompass different accountability and governance arrange-ments compared to traditional procurement – indeed, these arrangements areone of the claimed advantages of this provision method. Interlinked finan-cial incentives across a consortium of players, the sharing of risks throughcarefully contractualized legal relationships, and more flexible decision-making processes between executive government and service provider all feature as improvements over traditional procurement arrangements. The progressive contractualization of the state’s services and activities has beenaccompanied by the general assumption of increased accountability in all its forms, although this has rarely been tested. Whilst contractualization mayhave increased managerial accountability, it may have been at the expenseof reduced public accountability in its various forms.

Summing up, PPPs present challenges to the traditional forms of publicadministration in terms of the complexity of the deals and the number ofactors involved; the government’s capacity to steer; the choice between govern-ance forms; possible strategic behavior by the private-sector organizations;and the duration of contracts which makes evaluation difficult. In that way,PPPs share many of the same features associated with the emergence of thenew public governance as described in this book.

Discussion

These challenges remind us that it might take some time before PPPs arefully compatible within an elaborate public governance structure. The challenges point to a broader issue of the place of PPPs within a public governance framework. These issues have to do with the democratic aspectsof public governance. The danger is that, if the challenges are not addressed,the PPPs will be difficult to align with democratic concerns.

PPPs shift the responsibility for some of the traditional steps associatedwith public policy-making in a democracy. It is commonplace to distinguishbetween policy formation, decision-making, implementation and results.When a PPP is established, the private sector is given the responsibility toimplement the policy and to create results. Decision-making about PPPs ismainly a technical matter, once the decision to propose a partnership optionhas been taken. Governments are often required to oversee a competitive

bidding round for the contract before deciding who can become the privatepartner in the PPP. In Europe, choosing among the bids must follow the criteria set out in the European Union public procurement rules, includingthe special rules for “a competitive dialogue”. After the contract to becomea partner has been awarded, the government leaves the implementation responsibility to the partnership organization, often called the SpecialPurpose Vehicle. Through the contract that binds the partners together, the partnership organization is then responsible for achieving the desired result – for example, financing, building and operating/maintaining a newhospital, a primary school or a sports facility. The private sector is givenmore responsibility, and now for a longer period than through ordinary competitive contracts.

The follow-up in terms of democratic regulation is not in place in all countries that have introduced PPPs. Whilst we have instituted a “regu-latory state” of independent regulators, ombudsmen and audit review bodies in order to disperse power away from political quarters after the privatization of state businesses (Hodge 2004c), this has not yet occurred withPPP deals. They have continued to be essentially two-way government–business deals rather than also involving the community or any other independent accountability body to protect the public’s interests. They have also been handled on a case-by-case basis, by the government itself, in the face of multiple conflicts of interest. The potential for the interests of the advocating government and business partners to dominate the public interest is palpable here. For example, early drafts of the Australianstate of Victoria’s PPP guideline materials did not even mention the “publicinterest” notion and treated government solely as if it were a contractual partner in a commercial deal. In many ways this is reminiscent of past centuries.

Communities seem to need far more discussion and debate as to how theymight better ensure that the public interest is met through PPP deals, as well as meeting the needs of the contracting parties. To the extent that newinfrastructure contract delivery arrangements have reduced existing account-ability arrangements and altered longstanding governance assumptionswithout democratic debate, new partnership arrangements lack legitimacy(Hodge 2006).

A key point is if the criticisms aimed at PPPs can be remedied in the longrun. Or is there something inherently difficult in aspiring toward a “genuine”partnership between public-sector organizations and private organizations?One view could be that it is just a matter of more analysis needed which willimprove the basis for decision-making and eventually secure the potentialbenefits (US Government Accountabiliy Office 2008). Recalling the old saying that the “public and private management is alike in all unimportantaspects” (see the discussion in Alison 1992), the question is if the two

sectors can genuinely work together or if the competing interests are simplytoo great to overcome. Money is the end of private business while money isa means in the public sector to achieve a social outcome (Moore 2000). Itcannot be denied that a lot of the literature on PPPs in the beginning wasof a hopeful nature that emphasized the opportunities of cooperation. In recentyears, as we acquire more empirical analysis on PPPs, the verdict becomesmore nuanced. A proper way forward for research on PPPs is to rely evenmore on empirical investigation that can bring fresh insights to the questionof whether PPPs work in practice and how they fit into the institutions of public governance and the wider democratic context in which they areembedded.

It would seem that these aspects of democratic governance should be investigated more in the PPP debate. So far much of the literature has beenpreoccupied with the performance of PPPs and the legal aspects of PPPs,but seems to pay less attention to the governance challenges faced bydemocracies if they were to use PPPs more systematically. Of course, manygovernments may not, in fact, make use of PPPs that much; and, if that is the case, PPPs are confined to a special corner in the public governancediscussion.

The global financial crisis that occurred in the late 2000s sparked off anew round of debate as to how PPPs are appropriate in a new public governance framework. The promise of transfer of knowledge and know-how from the private partner to public organizations is probably still an attraction to governments around the world (Moszoro and Gassorowski 2008).But the uncertainty connected to the long-term impact of the financial crisis may cast a shadow over the future attractiveness of the PPP model.One of the first reports on the topic estimated that the global financial crisis offers both governments and the private sector a chance to review therisk allocation schemes (Regan 2008).

Conclusion

PPPs for infrastructure projects came on to the policy scene in a big wayfrom the 1990s onward. The UK and Australia were among the first countries to adopt PPPs, although PPPs had been known in North Americaand other places under different names in earlier times. PPPs offered whatthe privatization policy before could not offer: close cooperation betweenpublic-sector partners and private-sector partners, risk-sharing, and long-termcontracts for infrastructure projects.

PPPs are but one of several options in the various governance forms available. Much emphasis has been laid on “from public to private” whilethe real question is that governments often find that they can choose among governance forms. PPPs offer some qualities, but are not seen as uni-versally attractive as governments may find inside production or complete

outsourcing a better alternative. The shift or not to PPPs does not alwaysoccur automatically, as Williamson’s (1985) theory of transaction costs anti-cipates, but requires empirical analysis to see why governments in certaincountries chose or abandoned the PPP policy option.

PPPs have usually been discussed in terms of their economic and finan-cial benefits or drawbacks, and PPP debates are often dominated by legalexperts or financial experts. In this chapter, we have sought to highlight the governance challenges associated with PPPs. There are a number of challenges connected with using PPPs. These include the complexity of thedeals, the role of the ministry of finance or key government sponsor of PPPdeals, choice between governance forms, government steering capacity, andthe time question (long-term contracts). How well these challenges are metis likely to influence how the PPP model fits into a democratic framework.PPP means involvement of the private sector in public policy-making to a greater degree than before, and questions also concern the robustness and readiness of regulatory institutions to deal with PPPs. A key point ofdiscussion is if genuine partnerships are possible within the PPP model, orwhether the deeper interests between public-sector organizations and privateorganizations will remain an obstacle to greater collaboration. More PPP literature is focusing on empirical results which are likely to cast light onsome of the questions. The discussion on PPPs appears to be moving frompromises of PPPs to a more evidence-based discussion of what works andwhat does not work in practice. The improved empirical evidence is likelyto influence the policy dialogue on the place of PPPs in the future institu-tions of public governance.

References

Alison, Graham T. (1992) “Public and Private Management: Are They Funda-mentally Alike in All Unimportant Aspects?”, in Jay Shafritz and Albert Hyde (eds)Classics in Public Administration, Belmont, Calif.: Wadsworth.

Allen Consulting Group (2007) Performance of PPPs and Traditional Procurementin Australia, Final Report to Infrastructure Partnerships Australia, 30 November2007.

Bloomfield, Pamela, Westerling, David and Carey, Robert (1998) “Innovation andRisks in a Public–private Partnership: Financing and Construction of a CapitalProject in Massachusetts”, Public Productivity and Review, 21 (4): 460–71.

Boardman, Anthony, Poschmann, Finn and Vining, Aidan (2005) “North AmericanInfrastructure P3s: Examples and Lessons Learned”, in Graeme Hodge andCarsten Greve (eds) The Challenge of Public–private Partnerships: Learning fromInternational Experience, Cheltenham: Edward Elgar.

Bovaird, Tony (2004) “Public–private Partnerships in Western Europe and the US: New Growths from Old Roots”, in Abby Ghobadian, David Gallear,Nicholas O’Regan and Howard Viney (eds) Public–private Partnerships: Policy and Experience, Basingstoke: Palgrave Macmillan.

Bovaird, Tony (2006) “Developing New Forms of Partnership with the ‘Market’ inthe Procurement of Public Services”, Public Administration, 84 (1): 81–102.

Broadbent, Jane and Laughlin, Richard (1999) “The Private Finance Initiative:Clarification of a Future Research Agenda”, Financial Accountability and Manage-ment, 15 (2): 95–114.

Deloitte (2006) Closing the Infrastructure Gap: The Role of Public–private Partnerships,London: Deloitte.

English, Linda (2005) “Using Public–private Partnerships to Deliver Social Infra-structure: The Australian Experience”, in Graeme Hodge and Carsten Greve (eds) The Challenge of Public–private Partnerships: Learning from InternationalExperience, Cheltenham: Edward Elgar.

Flinders, Matthew (2005) “The Politics of Public–private Partnerships”, BritishJournal of Politics and International Relations, 7: 215–39.

Frumkin, Peter (2002) “Service Contracting with Non-profit and For-profit Pro-viders: On Preserving a Mixed Organizational Ecology”, in John Donahue and JosephNye (eds) Market-based Governance, Washington, DC: The Brookings Institution.

Ghobadian, Abby, Gallear, David, O’Regan, Nicholas and Viney, Howard (eds) (2004) Public–private Partnerships: Policy and Experience, Basingstoke: PalgraveMacmillan.

Heinrich, Carolyn and Lynn, Laurence (eds) (2001) Governance and Performance,Washington, DC: Georgetown University Press.

Her Majesty’s Treasury (2003) PFI: Meeting the Investment Challenge, London: TheStationery Office.

Her Majesty’s Treasury (2006) PFI: Strengthening Long Term Partnerships, London:The Stationery Office.

Her Majesty’s Treasury (2008) Infrastructure Procurement: Delivering Long-termValue, London: The Stationery Office.

Hodge, Graeme (2005) “Public–private partnerships: The Australian Experiencewith Physical Infrastructure”, in Graeme Hodge and Carsten Greve (eds) TheChallenge of Public–private Partnerships: Learning from International Experience,Cheltenham: Edward Elgar.

Hodge, Graeme (2006) “Public–private Partnerships and Legitimacy”, University ofNew South Wales Law Journal, Forum, 12 (2).

Hodge, Graeme and Greve, Carsten (eds) (2005) The Challenge of Public–privatePartnerships: Learning from International Experience, Cheltenham: Edward Elgar.

Hodge, Graeme and Greve, Carsten (2007) “Public–private Partnerships: An Inter-national Performance Review”, Public Administration Review, 67 (3): 545–58.

Institute of Public Policy Research (UK) (2001) Building Better Partnerships,London: IPPR.

Karmarck, Elaine (2007) The End of Government as We Know It: Making Public PolicyWork, Boulder, Colo.: Lynne Rienner.

Kettl, Donald F. (1993) Sharing Power, Washington, DC: The Brookings Institution.Klijn, Erik-Hans and Teisman, Geert (2005) “Public–private Partnerships as the

Management of Co-production: Strategic and Institutional Obstacles in a Diffi-cult Marriage”, in Graeme Hodge and Carsten Greve (eds) The Challenge ofPublic–private Partnerships: Learning from International Experience, Cheltenham:Edward Elgar.

Linder, Stephen (1999) “Coming to Terms with the Public–private Partnership: A Grammar of Multiple Meanings”, American Behavioral Scientist, 43 (1): 35–51.

Moore, Mark H. (2000) “Managing for Value: Organizational Strategy in For-profit, Nonprofit and Governmental Organizations”, Nonprofit and VoluntarySector Quarterly, 29 (1), supplement: 183–204.

Moszoro, Marian and Gasiorowski, Pawel (2008) The Optimal Capital Structure ofPublic–private Partnerships, International Monetary Fund Working Paper WP 08/1,Washington, DC: International Monetary Fund.

Mott Macdonald (2002) Review of Large Public Procurement in the UK, London: MottMacdonald.

National Audit Office (2003) PFI: Construction Performance, HC 371, London: TheStationery Office.

Organisation for Economic Cooperation and Development (OECD) (2008) Public–private Partnerships: In Pursuit of Risk Sharing and Value for Money, Paris:OECD.

Osborne, Stephen (ed.) (2001b) Public–private Partnerships: Theory and Practice inInternational Perspective, London: Routledge.

Pierson, Paul (2004) Politics in Time, Princeton, NJ: Princeton University Press.Pollitt, Christopher (2007) Governing with the Past, Oxford: Oxford University

Press.Pollitt, Michael (2005) “Learning from the UK Private Finance Initiative Experi-

ence”, in Graeme Hodge and Carsten Greve (eds) The Challenge of Public–privatePartnerships: Learning from International Experience, Cheltenham: Edward Elgar.

Pollock, Allyson, Price, David and Player, Stewart (2007) “An Examination of the UK Treasury’s Evidence Base for Cost and Time Overrun Data in UK Value-for-money Policy and Appraisal”, Public Money and Management, April:127–33.

PriceWaterhouseCoopers (2005) Delivering the PPP Promise: A Review of PPPIssues and Activities, available at www.pwc.com

Rosenau, Pauline Vaillancourt (ed.) (2000) Public–private Policy Partnerships,Cambridge, Mass.: MIT Press.

Shaoul, Jean (2005) “The Private Finance Initiative or the Public Funding of PrivateProfit”, in Graeme Hodge and Carsten Greve (eds) The Challenge of Public–private Partnerships: Learning from International Experience, Cheltenham: Edward Elgar.

Teisman, Geert and Klijn, Erik-Hans (2001) “Public–private Partnerships in theEuropean Union: Official Suspect, Embraced in Daily Practice”, in StephenOsborne (ed.) Public–Private Partnerships, London: Routledge.

Terry, Francis (1996) “The Private Finance Initiative: Overdue Reform or PolicyBreakthrough?”, Public Money and Management, January–March: 9–16.

United States Government Accountability Office (2008) Highways Public–privatePartnerships: More Rigorous Up-front Analysis Could Better Secure Potential Benefitsand Protect the Public Interest, GAO 08–44, Washington, DC: GovernmentAccountability Office.

Van Ham, J. C. and Koppenjan, Joop (2001) “Building Public–private Partnerships:Assessing and Managing Risks in Port Development”, Public Management Review,4 (1): 593–616.

Wettenhall, Roger (2003) “The Rhetoric and Reality of Public–private Partnerships”,Public Organisation Review: A Global Journal, 3: 77–107.

Wettenhall, Roger (2005) “The Public–private Interface: Surveying the History”, in Graeme Hodge and Carsten Greve (eds) The Challenge of Public–privatePartnerships: Learning from International Experience, Cheltenham: Edward Elgar.

Williamson, Oliver (1985) The Economic Institutions of Capitalism, New York: FreePress.

INTRODUCING THE THEORY OFCOLLABORATIVE ADVANTAGE

Siv Vangen and Chris Huxham

This chapter looks at some key governance tensions that are a consequenceof the increasingly interorganizational nature of public management. The focusis on the potential for collaborative advantage arising out of interorganiza-tional partnerships. We introduce the theory of collaborative advantage andexplain, through some illustrative examples, its ability to aid understandingabout the management of collaboration in practice.

A theme-based theory of collaboration

The theory of collaborative advantage was developed from research whichhas been ongoing since 1989, in which we have worked with and collecteddata from very many and varied types of collaborative situations, involvingindividuals with various roles including partnership managers, representativesof participating organizations and senior managers whose organizations areinvolved in collaboration. This has included supporting individuals who areaiming to drive particular collaborations forward, design and facilitation ofdevelopment events for collaborative partners, design and delivery of collab-orative leadership development events, contributions to practice seminars and conferences, direct participation in collaborations, both as participantsand as initiators and leaders, and contributing to policy development. The collaborations have ranged from “dyads” (two-party collaborations) tointernational worldwide networks. They have touched almost every aspectof the public and nonprofit sectors, and include public-private partnerships(PPPs) that also span the commercial sector. Their areas of concern take in, for example, a wide range of focuses on health and on education, anti-poverty, substance abuse, community development and planning, careers development, policing, economic development, and so on.

The theory has two organizing principles. First, it is structured around atension between Collaborative Advantage – the synergy that can be createdthrough joint working – and Collaborative Inertia – the tendency for collab-orative activities to be frustratingly slow to produce output or uncomfortablyconflict-ridden. It is also structured around issues that tend to energize those

who manage collaborations – their anxieties and rewards. It seeks to depictwhat underpins the anxiety and reward in each area. We call these issue areasthemes in collaboration practice (see Figure 10.1). Over the years, we havealso added topics that stem from our own observations, from policy con-siderations and from other researchers. The themes overlap with each other,so issues underlying each theme cross-relate with issues underlying others.

The theory itself is descriptive because it illustrates the complexity thatunderlies collaborative situations and the resulting challenges that areintrinsic to them. It is also prescriptive because it describes the issues thatmust be managed but without providing precise recipes for managerialaction. It thus recognizes the idiosyncratic nature of collaborative situationsand that there are positive and negative sides to alternative ways of managing.It provides conceptualizations and frameworks that can be used as “handlesfor reflective practice” (Huxham and Beech 2003) – that is, pointers to aspectsof collaborative situations that need deep consideration and management – thus enabling managers to think through and find ways of managing theirown collaborative situations.

This chapter provides brief overviews of four of these conceptualizationsand frameworks relating to agreement on aims, trust-building, culturaldiversity and attitudes to knowledge transfer. In so doing, we introduce someof the issues, tensions and challenges that generate collaborative inertia andsome ways for considering how to turn these around in order to gain theadvantage. In conclusion, we indicate how these four “themes” integrate withothers that we do not have space to cover here, to provide an overall senseof the nature of collaborative practice (see Huxham and Vangen 2005 for afuller holistic exposition of the theory of collaborative advantage). We alsointroduce some alternative conceptions of successful practice and discuss theirrelevance for understanding how to achieve collaborative advantage.

learningcommon

aimsidentity

culture

powercompromise

membershipstructures

workingprocesses

resources

commitment anddetermination

leadership democracyand

equality

communicationand language

accountability

social capitalrisk

Figure 10.1 Themes in collaboration practice.

distinction thus recognizes that aims at both the organizational and individuallevels motivate and influence the actions of those who enact the collaborationin practice.

(ii) Origin

The aims in the first dimension mostly relate to the concerns of members of the collaboration. However, aims formulated by members are sometimesstrongly influenced by the aims of organizations or individuals external tothe collaboration. Government is perhaps the most common organizationalstakeholder exerting pressure on collaborations, and it frequently influencesand shapes them. Whether collaborations are mandated or constrained bygovernment, nationwide policies as well as local priorities and interests tendto have an effect on the aims of the collaboration.

(iii) Authenticity

Aims expressed by members and external stakeholders may be genuinestatements about what they aspire to achieve. However, there are many reasons why members may not identify with aims that are nevertheless pub-licly stated. For example, they may not seriously subscribe to aims that havebeen imposed upon them by external pressure, or changes in the situationmay have altered the relevance of previously genuine aims. Organizationsmay, for example, invent a jointly owned substantive aim that satisfies thespecifications of a funding-provider and which effectively disguises their real aim. Similarly, individuals may invent aims for their organizations tolegitimize their own personal involvement in the collaboration. We charac-terize such possibilities as pseudo-aims.

(iv) Relevance

The identification of specific aims for each of the parties involved as well as the joint purpose is acknowledged as important if the collaboration is tosucceed. Recognizing which organizational aims can reasonably be pursuedthrough the collaboration is, however, not always straightforward. Otherrelated aims remain to be addressed by the organization alone or perhapsthrough other collaborations. The fourth dimension thus distinguishes thoseaims that should or are intended to relate specifically to the collaborativeagenda from those that are closely related but not explicitly a part of it.

(v) Content

Many of the aims expressed by individuals are essentially concerned with whatthe collaboration is about, such as gaining access to resource and expertise,

sharing risk, improving coordination in service provision and increasingefficiency. They relate to substantive outcomes and are obviously importantin all collaborations. However, participants also – often implicitly – expressaims that relate to how the collaboration will be undertaken. These aims canrelate to any aspect of collaborative processes, so might, for example, relateto modes of communicating, to the kind of relationship between membersor to a myriad other possibilities.

(vi) Overtness

Finally, aims may be openly discussed and explicitly stated, but there are alsomany reasons why they may knowingly not be revealed to other partici-pants, even if there is genuine goodwill between partners. Hidden agendasare endemic in collaboration. Deliberate concealing of aims is, however, not the only reason why they may not be clearly stated. In practice, theremay be limited opportunities to discuss explicitly all potentially relevant aimsin open forum; many aims go unstated even when there is no intent to hide them.

Taken together, these dimensions thus indicate that the aims relevant to collaborative situations will relate to aspirations not only for the collabor-ation but also for the organizations and individuals involved; may have beengenerated by those involved but may also have been imposed or suggestedby external stakeholders; may be genuine but can also be manufactured toprovide a reason for involvement in collaboration; do not always relate tothe activities of the collaboration; can relate to substantive or processual con-cerns; and do not all appear overtly in the discourse of the collaboration.

Conceptualizing the aims across these six dimensions may help facilitatebetter understanding of the variety of aims that are relevant to collaborations,and the ways in which multiple and conflicting aims can prevent agreementand progress. When faced with the task of managing collaborations, tryingto identify aims across the dimensions, as they may be perceived by all partners, can be very enlightening, whether it is done as a quick and easy ora more thorough investigative exercise. Such an exercise can, for example,reveal difference between aims at the organizational level, which has the potential to cause conflicts of interest and misunderstandings. Similarly, whilstit is not possible to know others’ hidden agendas, it is possible to speculatethat they may exist and even have a go at guessing what they may be. Thereare a multitude of other types of insights that may be gained from applyingthe aims dimensions to real collaborative situations. The type of insights thatmay thus be gained can be very helpful in understanding and judging howbest to work collaboratively with partners.

At the general level, the obvious conclusion to be drawn from this con-ceptualization is that agreeing on aims is frequently problematic in practice

INTRODUCING THE THEORY OF COLLABORATIVE ADVANTAGE

and that discussion aimed at establishing clarity of purpose can easily unearthirreconcilable differences. Arguably, it may be prudent to get started on sometangible actions without fully agreeing on aims. Instead partners should seek to gain enough agreement to allow them to make progress. Exactly what constitutes enough depends on partners’ assessment of the particularsituations. Time set aside for incremental collaborative achievements can help build trust between partners which can then form the basis upon whichdifficult discussions about aims can be held.

Managing trust

Trust is a prerequisite for successful collaboration yet many situations arecharacterized by suspicion and mistrust.

As with the issue of aims, trust is also seen as a necessary condition forsuccessful collaboration (Lane and Bachmann 1998) yet the reality of manycollaborations suggests that trust is frequently weak – if not lacking altogether. This particular paradox then suggests that there is a need to lookat how trust can be built and maintained between partners in the context of collaboration.

One way of conceptualizing trust-building is through the loop depicted in Figure 10.3. This argues that two factors are important in initiating a trust-ing relationship. The first concerns the formation of expectations about the

Reinforce trustingattitudes

Aim for realistic(initially modest)but successful

outcomes

Gain underpinningsfor more ambitious

collaboration

Form expectations aboutthe future of the

collaboration based onreputation or past

behavior or contractsand agreements

Have enough trust andtake a risk to initiate

the collaboration

Figure 10.3 The trust-building loop.

future of the collaboration; these will be based either on reputation or onpast behavior, or on more formal contracts and agreements (Gulati 1995).Given the difficulties of agreeing aims, as discussed above, this is a nontrivialstarting point. The second starting point involves risk-taking: partners need to trust each other enough to allow them to take a risk to initiate the collaboration (Gambetta 1988). If both of these initiators are possible,then, the loop argues, trust can be built gradually through starting with some modest but realistic aims that are likely to be successfully realized. This reinforces trusting attitudes and provides a basis for more ambitious collaboration.

This conceptualization of trust-building aligns itself well with a “small wins”approach (Bryson 1988) within which trust can be built through mutual experience of advantage gained via successful implementation of low-risk initiatives. Trust can be developed over time, moving gradually toward initiatives where partners are willing to take greater risks because a high level of trust is present. When risk and uncertainty levels are high, a strategyinvolving incremental increases in resource commitments may indeed be thepreferred strategy. In many situations, however, the collaborative advantageaimed for requires the collaborating partners to be more ambitious, and hence to adopt a higher-risk approach. The small-wins approach may, forexample, be in contradiction to the need to address major social issues rapidly or meet the requirements of external funding bodies for demon-strable output. More comprehensive ways of managing trust have different implications for initiating and sustaining the trust-building loop. We shallelaborate further on this below.

Initiating the trust-building loop

(a) Forming expectations

Two structural features – ambiguity and complexity – that tend to charac-terize collaborations can act as barriers to the initiation of trust-building.Whilst researchers have argued that “explicit” membership where the parties “know and agree on who is involved and in what capacity” is a keydefinitional element of collaboration, the surprising reality of many situ-ations is the ambiguity about who the partners are. Typically, there are dif-ferences in views about who the central members are and what their rolesor membership status are with respect to the collaboration. In practice it canbe difficult to be certain about what organization, collaboration or other con-stituency (if any) individuals represent. Simply identifying with whom to buildtrust therefore can be very difficult and time-consuming.

Working out with whom trust should be built is not the only challenge ingetting started in the trust-building loop. As we have already discussed above,practitioners continuously raise concerns over the establishment of joint aims.

Seeking agreement on aims to initiate the trust-building loop effectively canbe problematic in practice.

(b) Managing risk

Gradually, as trust develops, it becomes a means for dealing with risk. Insituations where the small-wins approach is not feasible, however, the riskassociated with the collaboration has to be managed as an integral part oftrust-building. Risk is usually associated with opportunistic behavior and vulnerability relating to apprehensions that partners will take advantage ofcollaborative efforts by, for example, claiming ownership of joint efforts. Whenthe aim is to build trust, however, risk management cannot be concernedwith guarding against opportunistic behavior, e.g. via sanctions set out in con-tractual agreements. Instead risk management must ensure that any futurecollaborative advantage can realistically be envisaged and is shared. Thisrequires efforts associated with aims negotiation, structural ambiguity, clarifica-tion of expectations, willingness and ability to enact the agreed collaborativeagenda in view of associated power and influence relationships, and so on.These activities are extremely resource-intensive and time-consuming, andtheir management requires a great deal of skill and sensitivity. Hence theeffort is only recommended where trust cannot be built incrementally.

Sustaining the trust-building loop

(c) Managing dynamics

Many collaborations are initiated, so it must be presumed that expectationscan be formed on the basis that either a minimal level of trust is present and/orthere is a willingness to bear the associated risk. Sustaining the trust-buildingloop then requires the participants to work together, gradually becoming more ambitious, over time, in their joint endeavors. Unfortunately, while all organizations are dynamic in nature, collaborations are particularly sobecause they are sensitive to transformation in each of the partner organiza-tions and therefore may change very quickly. Effort put into building mutualunderstanding and developing trust can be shattered, for example, by a changein the structure of a key organization or the job-change of a key individual.Sustaining the trust-building loop therefore requires continuous attention totrust relationships.

(d) Managing power imbalances

Even when careful and continuous attention is paid to trust-building, the inher-ent fragility of the loop is evident. Alongside the issues relating to the dynamicnature of collaboration, power issues in particular seem to challenge efforts

aimed at sustaining the loop. Imbalance in power, and the inevitability thatsome partners will be more central to the enactment of the collaborative agendathan others, tends to dictate behaviors that get in the way of trust-building.An appreciation of the inevitability of power imbalances as well as the ability to interpret any actions that members take in response to them mayhelp prevent loss of trust. Furthermore, an understanding of the way in whichbalances of power tend to change during the life of a collaboration, and indeedwhether and how power imbalances can and should be deliberately shifted,seems essential in sustaining the trust gained.

(e) Nurturing the collaborative relationships

Issues pertaining to the identification of partners, complexity and multiplicityof aims, risk and vulnerability, complexity and dynamics of collaborative struc-tures and power imbalances clearly all pose serious management challengesfor building and sustaining trust. If not managed effectively, any one of theseissues can prevent trust from developing or even cause loss of trust. Ideally,therefore, all these issues need to be managed simultaneously and, owing tothe dynamic nature of collaboration, in a continuous manner. Failing to doso may cause the trust loop to fracture.

This framework has sought to illustrate in broad terms the contrast betweentwo different approaches to the management of trust: small wins versus comprehensive management. Both approaches have their merits. The illus-tration of each intends to provide insight to inform the managerial judg-ment about the kind of trust-building activities that are appropriate to collaborative situations.

Managing culture

Cultural diversity is a source of advantage and inertia.We argued above that the bringing together of difference in terms of

partners’ resources, experiences and expertise provides the potential for col-laborative advantage. A significant part of this difference can be attributedto cultural diversity. In that sense, perceptions of difference are typically rootedin the national, organizational and professional cultures with which the various partners identify. A paradox arises because cultural diversity is seenboth as a source of stimulation and reward, and as a source of potentialconflicts of values, behaviors and beliefs (Bird and Osland 2006; Cray and Mallory 1998). The gist of the argument is that partners’ culturally embedded differences interact to cause tensions. These tensions can be con-ceptualized as pertaining to three areas in particular: encountering otherness,interaction order and harnessing difference (see Vangen and Winchester 2007 for a fuller elaboration on culture in collaboration).

(i) Encountering otherness

Encountering otherness involves the integration of different ways of being,interacting and working that are akin to specific cultures and which affectindividuals’ orientation toward others in the collaboration. Expectations of others’ behaviors tend to be based on perception of similarity – or evenstereotyping – of individuals within a culture. These expectations, research sug-gests, tend to lead to misaligned expectations and perceptions of superioritypotentially affecting every strategic and operational aspect of a collaboration(see e.g. Bird and Osland 2006; Walsh 2004).

Encountering otherness, therefore, gives rise to a particular managerial tension. On the one hand, it is necessary to build cultural awareness to tacklemisaligned expectations and perceptions of superiority and, on the other hand,it is essential to avoid the pitfalls of stereotyping. The tension arises becausethe act of building cultural awareness necessarily implies a need to con-ceptualize and generalize; thus, the danger of stereotyping is inherent in the learning. In any event, as each collaborative situation is unique in its configuration of cultural communities of belonging, the generic learning can-not capture the specific confluence of cultural diversity within each specificsetting. Hence the requisite understanding is not necessarily transferable from one situation to another.

Working effectively with otherness involves knowing how to manage theidiosyncratic nature of each individual situation. Yet individuals will enternew situations with embedded “ways of being” which may not feasibly berevisited with every new or significantly changed collaborative situation.

(ii) Interaction order

Whilst encountering otherness is concerned with individuals’ embedded perceptions and expectation of their partners, other cultural differences are manifested in the working practices and interaction between partners. We use the term “interaction order” to refer broadly to partners’ workingpractices.

The most important aspect is communication. Most obviously, the con-fluence of different natural and professional languages and organization-specificjargon increases the tendency for misinterpretations and misunderstandings.Furthermore, culturally determined etiquette generates issues concernedwith initiating, managing the content and managing the style of communi-cation. In terms of initiation, partners frequently have different views aboutwhether formal titles should be used or not, what formal level or organ-izational position individuals need to hold to represent their organization in communication with their partners, and so on. Similarly, the content ofcommunication can be influenced by differences in the formality of a culture

in terms of, for example, whether specific guidelines are required or not. Lastly, styles of communicating vary across cultures, so that some partners,for example, prefer to be succinct and to-the-point whereas others like toengage in lengthy negotiations.

The need to establish an interaction order thus points to a particular man-agement tension; cultural sensitivity is necessary to interact across differentcommunities yet a generic form of communication is necessary to enact thejoint agenda. The tension arises because different cultures encompass differ-ent natural and professional languages and culturally determined etiquettes,which impact on both the process and the content of communication.Paying attention to these differences may be essential in securing effectivecommunication. At the same time, any generic form of communication, necessary to enact the joint agenda, may not accommodate all the culturaldiversity present in any specific collaboration.

(iii) Harnessing difference

Harnessing individuals’ embedded perceptions, behavioral characteristicsand professional expertise alongside differences in organizations’ systems and procedures is essential if the advantage is to be gained. This points toa further three inter-related managerial tensions:

(A) FLEXIBILITY VERSUS ESTABLISHED ORGANIZATIONAL PROCEDURES

Flexibility – at the individual and organizational level – is necessary to accom-modate the encounterment of otherness and differences in interaction order.Organizations’ structures and procedures, for example, reflect the nature oftheir individual remits, and their ability to collaborate is associated with theirability to adapt to the needs of the joint agenda. Yet organizations’ contribu-tion to the collaboration is typically rooted in their unique resources andexpertise, so that any alteration – to structures and procedures – is done atthe risk of upsetting those that work. Thus, whilst flexibility is necessary toenact the collaborative agenda, this is most likely in tension with retainingthose established procedures that enable organizations to make a contribu-tion in the first place.

(B) AUTONOMY VERSUS ACCOUNTABILITY

In many situations, then, harnessing difference entails working within systemsthat are ill-fit for the purpose. Individual managers typically lack discretion,power and authority to act on behalf of their organization. This can makeit difficult for them to employ appropriately their understanding of their partner’s culturally embedded perceptions, behavioral characteristics and

professional expertise. This tension between autonomy and accountabilitycan play out in several ways. To harness differences effectively, individualsundoubtedly need enough autonomy to act on behalf of their organizations;yet at the same time they need to protect their organizations’ interest. Man-agers may thus end up compromising either their need to accommodate cultural difference at the level of the collaboration or their accountabilityfor their actions back to their organizations.

(C) RETAINING CONTROL VERSUS GRAPPLING WITH COMPLEXITY

Working within ill-fitting structures and procedures with insufficient amountsof autonomy will not render the accommodation of cultural differences simple. The greater the diversity of partners involved, the greater the complexity that needs to be managed. Two common responses to handling complexity are to seek partners where there is a greater degree of similarityor to control, via communication, the activities of the collaboration as a whole. There is a real opportunity cost associated with simplifying culturaldiversity in this way. The potential strength, and hence the real potential for collaborative advantage, rests on the collaboration’s ability to tap into the expertise and experiences of all involved. Limiting the number ofindividuals involved in communication, for example, implies that the potential for collaborative advantage is limited to the strength of commu-nication between those individuals. There is thus a real tension in dealingwith the complexities stemming from the number of stakeholders that areinvolved: whilst retaining control is a necessary element of steering the joint agenda forward, grappling with complexity is necessary if the collabor-ation is to harness the expertise of those who hold the potential to creatingadvantage.

The tensions that we have conceptualized seek to illustrate that cultural differences – whether rooted in nations, organizations or professions – can impact on all aspects of collaboration. Successfully managing cultural diversity involves paying careful attention to the plurality of issues arisingfrom cultural interaction. The managerial tensions outlined in this section,as summarized in Figure 10.4, define key areas where attention is required.

Managing knowledge transfer

Attitudes to knowledge-sharing vary.While there may be many reasons for collaborating, “interorganizational

learning” is a key aspect of all (Hartley and Allison 2002). For example, inthe case of a collaboration over care of the elderly between a health agencyand a social services department, social services staff might learn how to carry out some of the care tasks previously carried out by health staff, andhealth staff might learn how to advise patients about possible state benefits

Harnessing difference

Flexibilityiii) Established OrganizationalProcedures

VersusFlexibility in structures andprocesses is necessary toaccommodate the encountermentof otherness and differences ininteraction order.

Established structures andprocedures may be what enableorganizations to make acontribution to the joint agenda inthe first place.

Interaction order

Cultural Sensitivityii) Generic Communication

VersusCultural sensitivity is necessaryto effectively interact acrossdifferent communities.

Any specific form of communication,necessary to enact the jointagenda, may not accommodate allthe cultural diversity present in anyspecific collaboration.

Retaining Controlv) Grappling with Complexity

VersusRetaining control is a necessaryelement of steering the jointagenda forward.

Grappling with complexity isnecessary if the collaboration isto harness the expertise of thosewho hold the potential to creatingadvantage.

Autonomyiv) Accountability

VersusIndividuals need autonomy to acton behalf of their organizations totake the agenda forward.

Accountability is necessary toprotect organizations’ interestsand their inherent contribution tothe collaboration.

Encountering otherness

MANAGERIAL TENSIONS

Misaligned Expectationsi) Sophisticated Stereotyping

VersusLearning aimed at building culturalawareness is necessary to tacklemisaligned expectations andperceptions of superiority.

Generic learning cannot capturethe specific confluence of culturaldiversity within a specific setting.The act of building culturalawareness necessarily impliesa need to conceptualize andgeneralize, inherently incorporatingthe pitfalls of sophisticatedstereotyping.

Figure 10.4 Tensions in the management of culture.

for which they might be eligible. “Substantive learning” (Huxham and Hibbert2008) of this type generally relates to knowledge transfer where knowledgeembedded in one organization eventually also becomes embedded in a partnerorganization. Knowledge transfer may be an explicit reason (or part of thereason) for setting up a collaboration. For example, an aim of a communitysubstance-abuse partnership might be that police and health workers learnto understand each other’s perspective and expertise on the issue. More oftenthan not, however, knowledge transfer happens alongside pursuit of the explicitreason for collaborating. Substantive learning may also, however, relate toknowledge creation in which the partners jointly work together to create some-thing that neither previously had. Collaborative knowledge creation is animportant concept in the context of innovation management.

Conventional accounts of interorganizational learning tend to constructit in one of three ways. Sometimes partners are construed as having a selfish attitude, acquiring knowledge from a partner exclusively for their own use in an exploitative manner (Ingram 2002). Sometimes they are seenas having a sharing attitude, which leads them either to exchange knowledgewith partners or to go further and explore innovative solutions to problems at hand collaboratively (Inkpen and Tsang 2005). Quite often, however, no consideration is given to substantive learning at all, in which case the attitude to it is essentially one of sidelining, although learning is very likelyto happen alongside whatever other activities the collaboration holds(Spekman, Isabella and MacAvoy 2000). These stereotypical attitudes canbe characterized as follows:

Selfish: we take from you without giving to you;Sharing – exchanging: we take from you and we give to you; you take from

us and give to us;Sharing – exploring: we take from you and we give to you; you take from

us and give to us – and we learn together to create knowledge;

Sidelining: learning from or with partners is not something we think about.

The first three are characteristically active stances in which the partners aredeliberate about the attitude they take, but sidelining is inherently passivein nature.

These conventional, stereotypical views of learning do indicate somethingabout the nature of learning in collaboration. It certainly does includeselfish, sharing and sidelining elements. However, in reality it is more complex than that. To understand more about the attitudes to learning thatparticipants actually bring to collaborations, it is helpful to break them downinto stances toward the giving and taking of knowledge from a partner. Ineach case, there are many possibilities. We shall look at them in turn.

(i) Stances to giving knowledge

Here are two examples of giving stances:

Protective: we don’t trust you, therefore we don’t give to you;Unconfident: we don’t trust our knowledge, so we shall not give it to you.

In both cases, members of the organization seek to avoid giving away know-ledge and may therefore be viewed by partners as selfish. But in neither case is the motivation selfish. In the first case, the protectiveness is born out of fear that partners will appropriate the knowledge to the detriment of the organization – for example, by encroaching on their territory. It is a reactive stance. In the second case, the active refusal to give knowledgecomes from a lack of faith in the validity of the knowledge – some statisticaldata, for example.

However, stances to giving do not always embody an unwillingness to pass knowledge to partners. Sometimes participants give willingly and maytherefore appear to the receiver to have a sharing attitude. This may, how-ever, mask underlying motives that are less than altruistic. For example:

Force feeding: when we need you to have understanding, we give to you;Parenting: when we deem it to be good for you, we give to you.

The attitude behind the first might be considered as pro-actively selfish. Thereceiving partner is being manipulated for the benefit of the giving one. Thesecond may be seen as an active, positively helpful attitude toward supportingand developing a partner; but it is nevertheless instrumental and controlling,so there is an element of selfishness about it.

(ii) Stances to taking knowledge

Attitudes to taking knowledge are equally divergent. Here are two examples:

Careless: we didn’t intend to take from you but since we now have the knowledge we shall use it (without considering the consequences for you);

Discretionary: we take from you only if we choose to.

The first case might occur, for example, when potential partners share ideasin order to see whether they might form a joint bid to a funding agency. Ifthey decide to make the bid independently, one may use ideas generated byanother. In this type of situation, participants in the taking organization areacting reactively to an opportunity that unexpectedly presents itself. Therewas no original intent to act selfishly, but that is what happens reactively

and often unwittingly. The original motivation is not selfish, but the con-sequences of carelessness for the idea-giving organization may be severe. Thesecond case represents the situation when participants take an active stanceagainst taking knowledge from a partner for purely selfish reasons linked to an assessment of the value of the knowledge to their organization. This kind of stance gives no consideration to whether their having the profferedknowledge would help the partner.

(iii) Stances in combination

These example stances are a few of the many that occur in collaborative situ-ations; many more examples are provided in Figures 10.5 and 10.6. Eachpartner’s attitude to learning with or from their partners is made up of attitudes to giving and taking (which are sometimes sidelined in the sensethat they are enacted unconsciously). These can vary depending on the typeof knowledge that is being transferred or created and will vary over time as the relationship changes and develops. Obviously attitudes to knowledgetransfer are linked to the aims that partners bring to the collaboration, andtheir interpretation of each other’s attitudes will have an important bearingon their ability to develop trusting relationships. The important point formanaging collaboration is that participants’ behavior may appear selfish but not derive from selfish motivations; and, equally, many appear sharing but are actually derived from selfish motivations. Taking a stereotypical viewof attitudes in the way described at the start of this section can thus lead to misinterpreting partners’ behavior. Misunderstandings of partners’ cul-tural norms may add significantly to the potential for this. Gaining the skillto “read a partner correctly” is therefore an important part of managing collaboration.

Conclusion: understanding collaborative success

The above four example themes indicate some of the challenges inherent inachieving collaborative advantage. They each present a slice of the overallpicture – a perspective on the challenges of collaboration – and, as we have indicated, they interact, with each affecting the others. The theory ofcollaborative advantage is constructed around many more of these themes.We do not have space to describe them all here, but we can summarize themost important of them. In terms of power, for example, the issue is thatpower-sharing is important yet there are often both real and perceivedpower imbalances between partners that tend to have a negative impact on behaviors. Membership structures are frequently ambiguous, complex and dynamic, which render issues to do with agreement on aims, buildingtrust, managing power relationships and cultural diversity infinitely hard.Leadership is likely to be beyond the scope of any one individual to tackle,

start. Five perspectives on success are important (Huxham and Hibbert 2007).The first does indeed relate to substantive outcomes of the collaboration, butthese can be highly varied in type. They may relate to financial or resourcegains – for example, better use of public funds – improvements in serviceprovision or much smaller “products” such as raised awareness, training orlearning. They may be beneficial for the organizations, for the individualsinvolved in a collaborative initiative or for the clients or citizens at whomthey are targeted. They may be short- or long-term outcomes. And they arealmost always relative gains; relative to what might have happened withoutthe collaboration or to how well other organizations in similar situations aredoing, for example.

The second perspective on success relates to the process of collaborating.Sometimes a successful process is just seen as a means to a substantive end.However, good processes can also be in themselves viewed as a measure of success. Like successful substantive outcomes, processes that havebecome good vary in type considerably. In some cases, the mere act of usingrespectful language to each other marks a successful process development.At the other end of the scale, a collaboration may be working well at anadvanced level, being highly pro-active, making good decisions and takingjoint action.

The third perspective overlaps to some degree with the first two. Emergentmilestones are not targets planned from the start, but semi-serendipitousachievements that arise along the way. They are signals that the collabor-ation actually achieved something, however great or small. Also highly varied, they are located on a continuum of tangibility. At the least tangibleextreme, milestones relate to the resolution of process issues; for example,it might be the point at which partners start to take account of each other’sinterests. At the most tangible extreme, they relate to physical artifacts; forexample, the co-location of partners into the same building. Between thesetwo extremes, emergent milestones include the production of a joint reportor the holding of a jointly organized event for a wider audience. Emergentmilestones are important success-indicators partly because achieving majorfinal targets can take a very long time, and partly because they often indi-cate something that turned out to be more significant than would have beenexpected in advance. They are often good trigger-points for helping the collaboration to move on to greater things.

The fourth perspective is concerned with recognition. Recognition of a collaboration by those not involved – for example, by articles in the pressor receipt of an award – is clearly a good – independent and verifiable – indicator of substantive success. However, recognition can also be valued inits own right. For example, recognition often comes in the shape of a requestto help another collaboration to achieve similar process or substantive ends.Recognition of individuals in a collaboration, in whatever form, can be seenas evidence of the skills they bring to the collaborative team.

The final perspective on success relates to personal pride. Statements ofsuccess are often wrapped up in language that promotes a positive personalidentity for the speaker; people who claim success sometimes go so far as to portray themselves as heroes. Being in a position to take pride in “a job well done” and identify with the collaboration is an important form of success. As with the other perspectives, success personified through pride can be expressed in a variety of ways. An individual might claim positiveoutput from the collaboration as her own doing, or portray herself as piv-otally pro-active or as having important personal qualities such as willpoweror experience. The pride perspective signifies collaborative success throughpersonal fulfillment. In that sense, it is different from the other perspectivesbecause the main beneficiary is the speaker herself. However, pride-successcan be argued to be more than just a personal indicator. Reaching the pointwhere individuals are willing to acknowledge overtly – and even promote –their own role in a partnership is significant.

The five perspectives on success provide a basis for developing realistic,rather than idealistic, expectations of what success in collaboration mightlook like. They emphasize that what is achieved is not necessarily pre-dictable. They are also useful for demonstrating success, to stakeholders – participants, constituents, customers, staff or policy-makers – even whena substantive outcome is not (yet) forthcoming. Most usefully, the five perspectives can be used as a framework to facilitate development of collaborative practice. They suggest signifiers that it is appropriate to watchout for and be positive about, even within a context where success is lessthan absolute.

Summary: using the theory of collaborative advantage

In this chapter we have introduced the theory of collaborative advantagethrough providing overviews of a selection of conceptualizations and frame-works pertaining to the management of aims, trust, cultural diversity andknowledge transfer. Our aim was to explain its ability to aid understandingabout the management of collaboration in practice. The various concep-tualizations and frameworks are intended for use as handles for reflectivepractice, which assumes real experience of collaboration.

The theory conveys that managing collaborations is a highly complexendeavor. It prescribes – through a focus on themes – the kinds of issuesthat need attention. However, it does not tell the user which of these themesto focus on, when to do so and how to use the information captured in themto guide them in their management actions. Knowing how to use the theoryis in itself a matter for managerial judgment. Themes such as aims, trust and cultural diversity or attitudes to knowledge transfer provide sensible starting points for consideration of managerial action. However, it may beequally prudent first to seek to get a sense of what success might look like.

This can in turn help identify the themes that are likely to yield valuableinsights. Similarly, in using these conceptualizations and frameworks ashandles of reflective practice, the nature of success envisaged may influencedecisions about alternative ways of managing.

Acknowledgments

We wish to thank our colleagues who allowed us to use our joint work inthis chapter: Nik Winchester for his work with Siv on “culture”, Paul Hibbertfor his work with Chris on “learning” and “success”, and Pam Hearne forher work with Chris on “success”. Chris also wishes to thank the AdvancedInstitute for Management Research, grant number RES-331-25-0016.

References

Bird, A. and Osland, J. S. (2006) “Making Sense of Intercultural Collaboration”,International Studies of Management and Organisation, 35 (4): 115–32.

Bryson, J. (1988) “Strategic Planning: Big Wins and Small Wins”, Public Money andManagement, 8 (3): 11–15.

Cray, D., Mallory, G. R. (1998) “Making Sense of Managing Culture”, Inter-national Thomson Business Press.

Gambetta, D. (1988) “Can We Trust?”, in D. Gambetta (ed.) Trust: Making andBreaking Cooperative Relations, Oxford/New York: Blackwell.

Gulati, R. (1995) “Does Familiarity Breed Trust? The Implications of Repeated Ties for Contractual Choice in Alliances”, Academy of Management Journal, 38(1): 85–112.

Hartley, J. and Allison, M. (2002) “Good, Better, Best? Inter-organizational Learningin a Network of Local Authorities”, Public Management Review, 4: 101–18.

Hudson, B. (2004) “Analysing Network Partnerships”, Public Management Review,6 (1): 75–94.

Huxham, C. and Beech, N. (2003) “Contrary Prescriptions: Recognizing GoodPractice Tensions in Management”, Organization Studies, 24: 69–94.

Huxham, C. and Hibbert, P. (2007) “Hit or Myth?: Stories of CollaborativeSuccess”, Partnership Sourcing Ltd (ed.) Partnering for Profit, London: PSL.

Huxham, C. and Hibbert, P. (2008) “Manifested Attitudes: Intricacies of Inter-partner Learning in Collaboration”, Journal of Management Studies, 45: 502–9.

Huxham, C. and Vangen, S. (2005) Managing to Collaborate: The Theory andPractice of Collaborative Advantage, London: Routledge.

Ingram, P. (2002) “Interorganizational Learning”, in J. Baum (ed.) The BlackwellCompanion to Organizations, Oxford: Blackwell.

Inkpen, A. C. and Tsang, E. W. K. (2005) “Social Capital, Networks and Know-ledge Transfer”, Academy of Management Review, 30: 146–65.

Lane, C. and Bachmann, R. (1998) Trust Within and Between Organizations: Con-ceptual Issues and Empirical Applications, Oxford: Oxford University Press.

Oliver, C. (1990) “Determinants of Interorganizational Relationships: Integration and Future Directions”, Academy of Management Review, 15 (2): 241–65.

Spekman, R. E., Isabella, L. A. and MacAvoy, T. C. (2000) Alliance Competence:Maximizing the Value of Your Partnerships, New York: John Wiley.

Vangen, S. and Winchester N. J. (2007) “An Exploration of Culture in Cross-national Collaborations”, presented to the BAM annual conference, Warwick, 11–13September. Available from authors.

Walsh, K. (2004) “Interpreting the Impact of Culture on Structure”, Journal of AppliedBehavioral Science, 40 (3): 302–22.

Further reading

Cropper, S., Ebers, M., Huxham, C. and Ring, S. P. (2008) The Oxford Handbookof Inter-organizational Relations, Oxford: Oxford University Press.

Huxham, C. and Vangen, S. (2005) Managing to Collaborate: The Theory andPractice of Collaborative Advantage, London: Routledge.

RELATIONSHIP MARKETING,RELATIONAL CAPITAL AND

THE GOVERNANCE OF PUBLICSERVICES DELIVERY1

Stephen P. Osborne, Kate McLaughlin and Celine Chew

Marketing in public services – an opportunity missed?

As government and voluntary and community organizations (VCOs) increas-ingly work in partnership to make public policy, and to design and deliverpublic services, new challenges are raised for the practice of marketing withinpublic service organizations (PSOs)2 and for assessing its fitness for purposewithin the “new regulatory state” (Jayasuriya 2004). However, this task iscomplicated by the considerable ambiguity surrounding the role of market-ing within PSOs. Indeed, the very existence of a “marketing function” withinPSOs is contentious in its own right. Based on English experience, this chap-ter seeks to develop clear conceptual foundations to guide the placing andpractice of this functional discipline within such organizations.

It argues that marketing activity has been embraced by PSOs, albeit with considerable reluctance, as an inevitable corollary to the NPM reformagenda of the past thirty years – with its dominant notions of consumerismand the “marketization” of public services, and rather than as a desirablemanagement discipline in its own right (Walsh 1994; Burton 1999). In doing so, it has neglected the opportunities offered for its development byalternative conceptualizations of marketing that take institutions, networksand governance, rather than consumers, as the central unit of analysis.

We argue that public services marketing practice over this period has been dominated by transactional models of marketing (Laing 2003) that havebelied the growing relational complexity of the above trends. Nowhere is there any evidence of a willingness to test the suitability and/or robustness of alternative models of marketing that might meet the needs of PSOs operat-ing within the current plural public policy environment, with its emphasison partnerships and relational contracting and governance (Erridge and Greer2002; Schwartz 2005; Bovaird 2006).

This chapter is in three parts and seeks to address this gap by exploringthe utility of precisely such an alternative model for practice (relationship marketing) by situating this within a broader model of relationship managementthat encompasses the associated concept of relational capital. It commencesby arguing that the model that has underpinned the development of publicservices marketing to date is an inappropriate one. Second, it validates thisargument by presenting three “micro-cameos” of the recent application of marketing activity to public services management. Third, it introduces the alternative model of relationship marketing and evaluates what such an alternative framework might contribute to the practice of marketing within PSOs.

Modeling PSO marketing behavior: the genesis of “public-sector marketing”

In the 1960s, marketing was predominantly presented as the key managerialdiscipline across all sectors of society. For example, Kotler and Levy (1969)argued that marketing was

. . . a pervasive societal activity [and that all organizations] are concerned about their “products” in the eyes of certain “consumers”and are seeking to find tools for furthering their acceptance.

(Kotler and Levy 1969: 10–12)

This position has subsequently been assumed by many marketing scholars and practitioners, and has become the starting point for drawing marketinginto the public domain – such as through social change and public educa-tion programs (social marketing), and debates about the rationing of publicservices (demarketing) and about the acquisition of resources by PSOs throughfundraising (Kotler and Zaltman 1971; Kotler and Andreason 1975; Laingand McKee 2000; Donaldson and O’Toole 2002).

However, this predominant perspective has increasingly been subject toan evolving ambiguity about the boundaries of marketing for PSOs. Havingargued previously for marketing as a generic organizational function, keyadvocates have subsequently shifted their ground to introduce an elementof marketing conditionality into the debate. In this perspective, five assump-tions are held to constrain the universality of the marketing concept:

• there are at least two parties to the exchange;• each party has something of value to offer to the other party;• each party is capable of communication and delivery;• each party is free to accept or reject the offer; and• each party believes it is appropriate or desirable to deal with the other

party (Kotler and Keller 2005).

These five conditions challenge the applicability of marketing to many public-service contexts. It is argued here that PSOs are not always free toaccept or reject public-policy initiatives instigated by politicians, for example,nor can their clients/users always be free to accept or reject a particular public-service offering (social control and primary/secondary education services being the most obvious examples of this).

Crucially, these early ideas about the conditionality of marketing exchangeare rooted in classical economics and a model of exchange theory that sup-ports a purely transactional view of marketing, with the firm as a unitaryentity operating in isolation from other organizations. This perspective onmarketing is based upon a model of discrete transactions that have “a dis-tinct beginning, short duration and sharp ending by performance” (Morganand Hunt 1994: 1). However, it has been criticized latterly in the manage-ment literature through the concepts of the “new competition” (Best 1993)and “new institutionalism” (Powell and DiMaggio 1991). Drawing upon thework of Benson (1975) on resource-dependency theory and Williamson (1985)on transaction-cost analysis, these new conceptualizations posit a model ofnetwork-based organizations that paradoxically compete by collaborating with other organizations, in order to lever in information, resources and capabilities. Such a model introduces new levels of complexity to exchangerelationships that are perhaps beyond the scope and competencies of tradi-tional transactional models of marketing and which, it is argued here, arehighly relevant to the practice of marketing for PSOs operating within theNew Public Governance regime.

The marketing function and PSOs in England

This has a comparatively recent history. It was not until the 1980s, when the“marketization” of public services under the then Conservative govern-ment commenced in earnest, that attention was turned to the potentialbenefits of marketing for PSOs operating in market and quasi-market conditions (Le Grand 1993; Scrivens 1991; Sheaff 1991; Walsh 1991). Thismarketization was thus a core driver for the growth of the marketing function within British PSOs. As such it has had a profound influence uponits trajectory since then. Central to this influence was the classical economicprinciple of exchange discussed above, which assumes market relations tobe based upon discrete and autonomous transactions and with little thoughtto the governance of reciprocal or ongoing transactions (Hindmoor 1998;Walsh 1991).

As a consequence of these assumptions, the marketing behavior of individual PSOs has invariably been highly individualistic in nature and self-seeking, arguably to the detriment of the whole public service system.Further, individual PSOs, and especially VCOs, have often been drawn into market scenarios for public services that have privileged competition and

RELATIONSHIP MARKETING AND RELATIONAL CAPITAL

adversarial relationships over collaboration both between PSOs and withinthem (Palmer 2001).

Three “micro-cameo” examples will illustrate this argument. First, the creation of the internal market in healthcare in the 1990s created com-petition between newly created “Trust” hospitals (service providers) and the primary care referral agents, usually general practitioners (service pur-chasers). The response of many individual Trust hospitals to this was to look to the marketing function to develop their competitive position withinthese newly created markets (Kitchener and Whipp 1995). The archetypal“4Ps” model, where organizations sought to strengthen their competitive position by varying dimensions of the product, its promotion, price and placeof distribution (Kearsey and Varey 1998; Sheaff 1991), informed marketingpractice over this period. As a consequence, individualism, interhospital rivalry and competitive behavior flourished in many areas. Individual Trust hospitals succeeded by pursuing their own corporate objectives andfinancial security – but at the cost of scant attention being paid to overallhealthcare policy or the holistic health needs of the local community(Fillingham 1994).

As a result of such behavior, marketing as a profession within PSOs became subject to damning critiques that it had led to goal displacement andstrategic drift in healthcare (Sheaff 1991; Scrivens 1991; Walsh 1994). Howcould either local or central government steer health policy, it was argued,when individual healthcare actors were more preoccupied with their own survival than with the health outcomes of patients?

A second example can be found in the field of the personal social services.The NHS and Community Care Act 1991 required local-authority SocialServices Departments to create a “mixed economy of care” for their services(Wistow et al. 1994). In their early attempts to create this new mixed economy, these departments saw VCOs as a natural choice as preferredproviders – both because of the perception of them as sympathetic to theneeds of vulnerable people (Brenton 1985) and because of their perceived(though unproven) institutional advantages, including cost efficiency, flexi-bility and consumer responsiveness (Knapp et al. 1990).

As a consequence, relationships between local government and VCOs weretransformed. The previous “grant-funded” relationship between local govern-ment and the VCO sector was replaced by a performance-based “contractculture” and the development of, often fractured, principal–agent relation-ships. In 1996 the Deakin Commission (Commission on the Future of theVoluntary Sector 1996) emphasized that a serious breakdown in relationsbetween the governmental and VCO sectors in England had resulted fromthis adversarial model and needed to be addressed by a shift into more explicitly relational approaches – leading to the development of the VoluntarySector Compact and later to Compact Plus (Osborne and McLaughlin 2002;Osborne 2003). This represented an emergent policy framework for building

relationships across organizations that emphasized inter- rather than intra-organizational management and governance.

As a result of this debate, some initial attempts were made to develop and test alternative models of the marketing function in PSOs (for example,Laing and McKee 2000; Laing and Hogg 2002; Wright and Taylor 2005).Notwithstanding these, though, the discrete transactional model of market-ing has continued to dominate practice within PSOs.

Finally, clients of the social security system in England are mandatory “customers” of the Benefits Agency – no alternative provider exists. As such,the Agency has been encouraged by central government to promulgate a “customer focus” as its dominant paradigm of behavior. In addition, though,it is also required by government to implement competing policies that challenge the notion of “claimants as customers” (Pheysey 1993).

Specifically, the alternative concerns of central government – to tackle theperceived negative impact of the benefit system on the work incentive andto minimize fraudulent claims – have produced a competing discourse ofclaimant behavior, as “workshy” and “fraudulent claimant” respectively. Inthis instance, therefore, the preferred marketing model of the agency, whichwas attempting to design consumerist choice into the system, was challengedby this alternative policy trajectory. Consequently, the limited transactionalmodel of the marketing function employed by the Agency was unable to copewith this level of policy complexity and paradox (Falconer and Ross 1999).

These micro-cameos throw up two important challenges for the future of the marketing function in PSOs. First, that its originating context of trans-actional and operational management has constrained the development ofmarketing within PSOs to a focus on intraorganizational and operational debates. This has crowded out consideration of more sophisticated models ofit as a basis for exploring its possible role in shaping the strategic behaviorof PSOs.

Second, that the place of marketing within PSOs has become a highly contested one. In VCOs, for example, the remit of the marketing functioncan often be limited to fundraising alone, rather than integrated into theirstrategic positioning activity (Chew 2003, 2006). Similarly, inside governmentalorganizations, marketing professionals have often had to “sail under anotherbanner” in order to have an impact on strategic direction and positioning(Piercy and Cravens 1995; Laing and McKee 2000).

It is contended here that the conventional transactional notion of market-ing, above, fails to accommodate such policy and system complexity, andoffers an incomplete basis for guiding marketing behavior within PSOs. Whatis required, therefore, is new, policy-relevant models of marketing that can assist PSOs in developing strategic responses to the growing relationalcomplexity of implementing public policy in the plural state. This returns usto the relational approaches to marketing and organizational managementnow evolving – but that to date have been little applied to PSOs.

Relationship marketing

Within the broader marketing literature it is increasingly acknowledged thatrelationship marketing (RM) represents an archetypal shift for marketing prac-titioners (Payne and Ballantyne 1993; Sheth and Parvatiyar 2000; Veloutsouet al. 2002). This shift acknowledges that sustainable competitive advantageincreasingly requires collaborative activity rather than rivalrous competition,as discussed above, and that relationships are often the most valuable resourceof a firm (Sharma and Patterson 1999; Helfert et al. 2002). This is the coreof RM, which Harker (1999: 16) has defined as an organization engaging in

. . . proactively creating, developing, and maintaining committed, inter-active and profitable exchanges with selected customers over time.

Gronroos (1994, 1999), in seminal papers, has argued that a marketing strategy continuum exists. At one end of this continuum is transactional marketing, rooted in classical economics, and that deals with one trans-action at a time. At the other end is RM, which focuses upon building relationships. The core of this relationship-building is trust. Drawing upontransactional economics, Selnes (1998: 308) notes that

. . . the importance of trust comes about . . . because of the diffi-culty or impossibility of acquiring information about future eventsor defining a contract that covers such future events.

Other authors have defined trust as a “willingness to rely on an exchangepartner in whom one has confidence” (Moorman et al. 1993) and have, crucially, specified its two core dimensions:

• trust in the honesty of a partner to fulfill obligations; and• trust in the benevolence of a partner to be genuinely interested in your

welfare, to be motivated to seek joint gains and not to seek opportunisticbenefit.3

Trust is argued to be at the core of RM. It both facilitates the adaptationprocess that is often necessary to complete an exchange within an ongoingrelationship (Brennan and Turnbull 1999) and provides the basis for mutualcommitment that “reduces the uncertainties associated with opportunisticbehaviour” in a volatile environment or market (Sheth et al. 2000). Giventhe present emphasis in public policy upon trust as a governance mechanismwithin the evolving plural state (Davis and Walker 1997; Osborne 2006), itis surprising therefore that RM has made only a marginal contribution tomarketing practice and organizational governance for PSOs. This chapternow presents the case for a more significant contribution.

Building the contribution of RM to public-policyimplementation and public-services delivery

Within the confines of this brief chapter, it is not possible to undertake acomprehensive exploration of the potential contributions of RM to publicmanagement and public governance. However, examples of three specificbenefits will suffice for our purpose.

Performing in the plural and pluralist state

In the introduction to this volume, Osborne argued that the dawn of thetwenty-first century has seen the evolution of the “new public governance”in public management, where the negotiation of both plural (involving multiple actors) and pluralist (involving multiple processes) public policy making and implementation is the dominant paradigm. Further, he has arguedthat traditional approaches to public policy implementation and public services delivery have failed to provide substantive guidance to public managers struggling to cope with this level of complexity – public adminis-tration invariably relegates the actual implementation of public policy to a“black box” that is subservient to the greater task of policy formulation whilstpublic management portrays the policy process as simply the context for theactual practice of public management. Both views are partial and flawed. RMoffers PSOs and public service managers an opportunity to move beyondthis simple bipolar opposition and to engage with the policy process in a waythat enhances pro-active interorganizational management and governance.

Morgan and Hunt (1994) have argued that RM actually comprises threelevels of marketing activity:

• the micro, organization–consumer, level (Berry 1983);• the macro, organization–organization, level (Arndt 1983); and• the meso, organization–society, level (Moorman et al. 1993).

This approach can offer PSOs a conceptual framework to move to a newplane of marketing sophistication. The micro-level concerns the explorationof co-production with the recipients of public services – both an essentialelement of good services management (Ravald and Groonroos 1996) and a core component of contemporary public policy in the UK and Europe(Brandsen and Pestoff 2006). The macro-level focuses attention upon bothboundary-spanning and boundary-maintenance activities for PSOs – bothessential for effective interorganizational collaboration (Kale et al. 2000; Tsai2000). Finally the meso-level involves the engagement of PSOs in the policyformulation and implementation process as cognizant, purposive actorsrather than as passive recipients (Ring and Van de Ven 1992; Gulati et al.2000). By conceptualizing marketing activity across these three levels, PSOs

can clarify its overall contribution to organizational performance. This is an essential first step in articulating a clear formulation of the place and role of the marketing function within PSOs and in replacing the current ambiguity and confusion.

Understanding and managing trust

Trust is at the heart of any interorganizational relationship and governance,including in public services provision (Osborne and Murray 2000; Huxhamand Vangen 2000). Often, though, “trust” appears in the literature as a reduc-tive self-defining concept. RM offers a different, dynamic perspective uponit, and one more suited to the challenges of PSOs within the plural state.The first of these challenges is the “principal–agent” problem (Vickers andYarrow 1988). This concerns the asymmetry of information that exists in relationships between two or more parties to a task. At its extreme, theprincipal to a partnership must employ a range of instruments in order tomonitor and control the behavior of their agent. Within a more relationalcontext, however, both can use their trust in each other to monitor the out-comes of their relationship rather than relying upon costly and bureaucraticperformance management systems, with all their implied transaction costs(Waterman and Meier 1998; Bachmann 2001). RM provides a clear basisupon which to build and maintain such trust – as well as warnings againstits misuse and potential lack of transparency (Morgan and Hunt 1994;Palmer et al. 2000).

The second challenge is that of dealing with risk, and its associated costs,in service innovation – a notoriously difficult activity in the risk-aversive culture of PSOs (Erridge and Greer 2002; Osborne and Brown 2005). A rela-tional approach based upon trust allows these risks and costs to be shared,minimizing the danger and sharing the benefits of a venture to each partyto the relationship. It also allows opportunistic behavior to be minimized,and the dedicated skills and knowledge base of an organization to be pro-tected (Lorenzoni and Lipparini 1999; Kale et al. 2000).

Thus, and third, trust is an input into relationship-building in the sensethat it is one of the core resources of any relationship. No ongoing relationshipwill survive without it. It is an output in the sense that working successfullytogether in a relationship reinforces and develops further the trust betweenthe parties involved – successful relationships breed deeper, and more success-ful, relationships (Ring and Van de Ven 1992; Gulati et al. 2000). Anotherkey to successful performance in the plural state, therefore, is the effectivegovernance of interorganizational relationships through mechanisms basedupon trust. RM offers a framework for PSO practitioners to understand andmediate these trust-based relationships.

Table 11.1 illustrates the challenge of moving toward such “relational governance”, when much of the marketing and managerial practice of the

recent decades has been based on a marketization model that has encour-aged discrete transactions and low trust levels between commissioners andproviders, as discussed above. Many current relationships are still bound by the classical and neo-classical governance of such individual transactions,whilst the policy trajectory, and the insights of RM, increasingly emphasizesrelational governance. At the core of this challenge, therefore, is how to makethe shift between these competing modes of governance. The concept of rela-tional capital, from the associated field of organizational strategy, providesfurther assistance here for PSO managers.

Relational capital – the heart of relationship governance

Donaldson and O’Toole (2002) argue persuasively that RM cannot standalone in the development and governance of productive interorganizationalrelationships. It has to go alongside other activity to optimize such relationships.One of the most significant concepts that the associated field of organiza-tional strategy can thus offer to such holistic relationship management by PSOsis that of “relational capital” (RC).

Kale et al. (2000) define RC as

. . . the level of mutual trust, respect and friendship that arises outof close interaction at the individual level between alliance partners.

(p. 218; our emphasis)

The key insight for PSO managers in this literature is to focus upon the import of individuals and individual relationships and their interaction with the organizational level of relationships. Too often it seems that, in the publicadministration and management field, relationships are reified to the organ-izational level alone – the neo-corporatist assumptions of the Voluntary Sector

Table 11.1 Commissioner–provider inter-relationships in public services provision(developed and adapted from the work of Ring and Van de Ven 1992: 490)

Interaction between service commissioner and service provider

Level of pre-existingtrust between service commissioner and service provider

Hierarchicalgovernance

Relationalgovernance

Market governance: discrete(classical) transactions

Market governance:recurrent (neo-classical)transactions governance

Compact in the UK are a good example of this (Osborne and McLaughlin2002). The RC approach does recognize the import of organizational factorsfor effective interorganizational working. However, it also makes explicit that these organizational factors are dependent upon key relationships at the individual level (where organizational staff interact with the staff of otherPSOs, policy-makers and service recipients) for this import.

The most recent research from the private-sector research has shown that 30–70 percent of all private-sector partnerships fail (Duysters et al. 1998; Park and Ungson 2001; Overby 2006) – and that RC is one of thestrongest safeguards against such failure (Morgan and Hunt 1994; Gulati et al. 2000). Thus, partnerships built upon RC become key strategicresources of an organization and enable it to gain a competitive advantageover its rivals not embedded in such partnerships (Kale et al. 2000; Tsai 2000; Sarkar et al. 2001). The key to deriving this advantage lies in the ways in which the key individuals in organizations learn to manage the balance between trust and distrust and conflict between their respectiveorganizations:

A firm derives its competitive strength from its proprietary assetsand will be protective about losing them to alliance partners.Partnerships are fraught with hidden agendas driven by the oppor-tunistic desire to access and internalize the partner’s core proprietaryskills. . . . [Relational capital creates] a mutual confidence that no party to an exchange will exploit others’ vulnerability even if thereis an opportunity to do so. . . . [This confidence] arises out of thesocial controls that [relational] capital creates.

(Kale et al. 2000: 222)

Finally, the private-sector literature also warns of the “dark side” of RC.Just as it can strengthen organizational performance, so it can undermineit, by tying in a firm to an unproductive or damaging relationship or by redu-cing its openness to new and potentially productive relationships (Parkhe and Miller 2000; De Weaver et al. 2005). As a consequence, it is essential tomanage the creation, sustenance and impact of RC upon an organization(Sarkar et al. 2001; Sawhney and Zabin 2001).

Conclusions

This chapter has argued that the practice of marketing in PSOs over recentdecades has been based within a classical model of marketing that hasemphasized both transactional relationships and rivalrous competition. Thishas been contrasted both with the current public policy trajectory towardthe governance of complex relationships and with best practice within theprivate sector that has emphasized the need for the governance of plural

relationships in order to perform most effectively in contemporary interorgan-izational fields. It has argued that the concepts of relationship marketing and relational capital have significant insights to offer to marketing and organizational management within PSOs, precisely because they examine thepractice and challenges of interorganizational management and governancein the plural state. It is not our case that “good marketing” is a substituteor proxy for “good management”. It certainly is not. But it does have animportant contribution to make to good governance and particularly to theextent that it can focus attention upon essential strategic decisions aboutinterorganizational relationships within the plural state.

Two insights are especially important. First, RM and RC offer a frame-work within which to locate and articulate the position of the marketing function within PSOs and that emphasizes its strategic rather than its operational significance. This should help clarify the current confusion andobfuscation about the positioning of this function. Second, they also offera framework for the practice of marketing in PSOs that emphasizes the challenges of the plural state and some guidelines as to intra- and inter-organizational management and governance within it. This chapter is a contribution to commencing a debate about these contributions as part of the development and critique of the public governance paradigm.

However, an important concluding caveat must be entered here. Almostall of the theory and evidence underpinning both RM and RC have beendeveloped from private-sector experience. The task now is to take these insightsto develop a model of relationship management that is more firmly rootedin public administration and management, and that offers real insight to PSOmanagers about the contribution that RM and RC can make to public services management. It will also need to acknowledge the limitations bothof these conceptual tools and of RM in the real world. No concept or toolis inherently positive. The limitations and dangers of RM and RC need to be explored as well as their positive contributions. Becoming too deeplyembedded within a relationship or network can risk being drawn into an unpro-ductive exchange or precluding the opportunity of developing new andalternative relationships, as discussed above.

The key task for PSO managers, and with which relationship marketingcan help, is thus to be able to make strategic choices about the extent andnature of organizational relationships and their import for organizationalperformance. Central to this is the place of relational capital in making areality of these relationships for staff and managers within PSOs, so that they can control and govern these relationships rather than be controlled by them. What the regime of public governance requires, it is suggested, isa new generation of relational capitalists who are adept not only in the core skills of their field or industry but also in facilitating, governing andsustaining the interorganizational relationships that are now at the heart ofpublic services provision.

Notes1 This chapter is a revised version of K. McLaughlin, S. Osborne and C. Chew (2009)

‘Relationship marketing, relational capital and the future of marketing in publicservice organization,’ in Public Money and Management, 29(1): 35–42.

2 Public Services Organizations are any organizations from across the governmental,VCO and business sectors that are involved in the provision of public services.

3 See also, inter alia, Morgan and Hunt 1994; Lorenzoni and Lipparini 1999; andGulati et al. 2000.

References

Arndt, J. (1983) “The Political Economy Paradigm: Foundation for Theory Buildingin Marketing”, Journal of Marketing, 47: 44–54.

Bachmann, R. (2001) “Trust, Power and Control in Trans-organizational Rela-tions”, Organization Studies, 22 (2): 337–65.

Benson, J. (1975) “The Interorganizational Network as a Political Economy”,Administrative Science Quarterly, 20: 229–49.

Berry, L. (1983) “Relationship Marketing”, in L. Berry, G. Shostack, and G. Upah(eds) Emerging Perspectives on Service Marketing, Chicago, Ill.: AMA.

Best, M. (1993) The New Competition, Cambridge, Mass.: Harvard University Press.Bovaird, T. (2006) “Developing New Forms of Partnership with the ‘Market’ in the

Procurement of Public Services”, Public Administration, 84 (1): 81–102.Brandsen, T. and Pestoff, V. (2006) “Co-production in the Third Sector and the Delivery

of Public Services”, Public Management Review, 8 (4): 493–502.Brennan, R. and Turnbull, P. (1999) “Adaptive Behavior in Buyer–seller Relation-

ships”, Industrial Marketing Management, 28 (5): 481–95.Brenton, M. (1985) The Voluntary Sector in British Social Services, Harlow:

Longman.Burton, S. (1999) “Marketing for Public Organizations: New Ways, New Methods”,

Public Management, 1 (3): 373–85.Chew, C. (2003) “What Factors Influence Positioning Strategies in Voluntary Non-

profit Organizations? Towards a Conceptual Framework”, Local Governance, 29 (4): 288–323.

Chew, C. (2006) “Positioning and Its Strategic Relevance: Emerging Themes fromthe Experiences of British Charitable Organizations”, Public Management Review,8 (2): 333–50.

Commission on the Future of the Voluntary Sector (1996) Meeting the Challenge ofChange: Voluntary Action into the 21st Century, London: Commission on the Futureof the Voluntary Sector.

Davis, H. and Walker, B. (1997) “Trust-based Relationships in Local GovernmentContracting”, Public Money and Management, 17 (4): 47–54.

De Weaver, S., Martens, R. and Vandenbempt, K. (2005) “The Impact of Trust onStrategic Resource Acquisition through Interorganizational Networks: Towards a Conceptual Model”, Human Relations, 58 (12): 1523–43.

Donaldson, B. and O’Toole, T. (2002) Strategic Market Relationships: From Strategyto Implementation, Chichester: John Wiley.

Duysters, G., De Man, A. and Wildeman, L. (1998) “A Network Approach to AllianceManagement”, European Management Journal, 17: 182–7.

Erridge, A. and Greer, J. (2002) “Partnerships and Public Procurement: Building SocialCapital through Supply Relations”, Public Administration, 80 (3): 503–22.

Falconer, P. and Ross, K. (1999) “Citizen’s Charters and Public Service Pro-vision: Lessons from the UK Experience”, International Review of AdministrativeSciences, 65 (3).

Fillingham, D. (1994) “Health in the Marketplace”, in H. Tam (ed.) MarketingCompetition and the Public Sector, Bath: Longman.

Gronroos, C. (1994) “From Marketing Mix to Relationship Marketing: Towards aParadigm Shift in Marketing”, Asia–Australia Marketing Journal, 2 (1): 9–29.

Gronroos, C. (1999) “Relationship Marketing: Challenges for the Organisation”, Journalof Business Research, 46 (3): 327–35.

Gulati, R., Nohria, N. and Zaheer, A. (2000) “Strategic Networks”, StrategicManagement Journal, 21 (3): 203–15.

Gutch, R., Kunz, C. and Spencer, K. (1990) Partners or Agents?, London: NCVO.Harker, M. (1999) “Relationship Marketing Defined? An Examination of Current

Relationship Marketing Definitions”, Marketing Intelligence Planning, 17 (1):13–20.

Helfert, G., Ritter, T. and Walter, A. (2002) “Redefining Market Orientation”,European Journal of Marketing, 36 (9–10): 1119–39.

Hindmoor, A. (1998) “The Importance of Being Trusted: Transaction Cost Theoryand Political Network Theory”, Public Administration, 76 (1).

Huxham, C. and Vangen, S. (2000) “Leadership in the Shaping and Implementationof Collaboration Agendas”, Academy of Management Journal, 4: 1159–77.

Jayasuriya, K. (2004) “The New Regulatory State and Relational Capital”, Policyand Politics, 32 (4): 487–501.

Kale, P., Singh, H. and Perlmutter, H. (2000) “Learning and Protection of ProprietaryAssets in Strategic Alliances”, Strategic Management Journal, 21 (3): 217–37.

Kearsey, A. and Varey, R. (1998) “Managerialist Thinking on Marketing for PublicServices”, Public Money and Management, 18 (April–June).

Kitchener, M. and Whipp. R. (1995) “Quality in the Marketing Change Process”, inM. Martinez and I. Kirkpatrick (eds) The Politics of Quality, London: Routledge.

Knapp, M. et al. (1990) “Public Money, Voluntary Action: Whose Welfare?”, in H. Anheier and W. Seibel (eds) The Third Sector, Berlin: De Gruyter.

Kotler, P. and Andreason, A. (1975) Strategic Marketing for Non-profit Organ-izations, Englewood Cliffs, NJ: Prentice Hall.

Kotler, P. and Keller, K. (2005) Marketing Management, New York: Prentice Hall.Kotler, P. and Levy, S. J. (1969) “Broadening the Concept of Marketing”, Journal

of Marketing, 33 (January): 10–15.Kotler, P. and Zaltman, G. (1971) “Social Marketing: An Approach to Planned Social

Change”, Journal of Marketing, 35: 3–12.Laing, A. (2003) “Marketing in the Public Sector: Toward a Typology of Public

Services”, Marketing Theory, 3 (4): 427–45.Laing, A. and Hogg, G. (2002) “Political Exhortation, Patient Expectation and

Professional Execution: Perspectives on the Consumerization of Health Care”, BritishJournal of Management, 13 (2): 173–88.

Laing, A. and McKee, L. (2000) “Structuring the Marketing Function in Profes-sional Service Organisations”, European Journal of Marketing, 34 (5–6): 576–97.

Le Grand, J. (1993) Quasi-markets and Social Policy, London: Macmillan.

Lorenzoni, G. and Lipparini, A. (1999) “The Leverage of Interfirm Relationships as a Distinctive Organizational Capability: A Longitudinal Study”, StrategicManagement Journal, 20 (4): 317–38.

Moorman, C., Deshpande, R. and Zaltman, G. (1993) “Factors Affecting Trust inMarket Research Relationships”, Journal of Marketing, 57 (1): 81–101.

Morgan, R. and Hunt, S. (1994) “The Commitment–trust Theory of RelationshipMarketing”, Journal of Marketing, 58: 20–38.

Osborne, S. (2003) “A Dance to the Music of Time: Evolving Relationships betweenGovernment and the Voluntary and Community Sector over Time”, LocalGovernance, 29 (4): 227–43.

Osborne, S. (2006) “The New Public Governance?”, Public Management Review, 8 (3):377–88.

Osborne, S. and Brown, K. (2005) Managing Change and Innovation in PublicService Organizations, London: Routlege.

Osborne, S. and McLaughlin, K. (2002) “Trends and Issues in the Implementationof Local ‘Voluntary Sector Compacts’ in England”, Public Money and Manage-ment, 22: 55–64.

Osborne, S. and Murray, V. (2000) “Understanding the Process of Public–privatePartnerships”, in S. Osborne (ed.) Public-Private Partnerships: Theory and Practicein International Perspective, London: Routledge, pp. 70–83.

Overby, S. (2006) “Big Deals, Big Savings, Big Problems”, CIO Magazine, 1 February.Palmer, A. (2001) “Co-operation and Collusion: Making the Distinction in Market-

ing Relationships’, Journal of Marketing Management, 17 (7–8): 761–84.Palmer, A., Barrett, S. and Ponsonby, S. (2000) “A Behavioural Analysis of

Co-operative Marketing Organisations”, Journal of Marketing Management, 16:273–90.

Park, S. and Ungson, G. (2001) “Interfirm Rivalry and Managerial Complexity: A Conceptual Framework of Alliance Failure”, Organization Science, 12: 37–53.

Parkhe, A. and Miller, S. (2000) “The Structure of Optimal Trust: A Comment andSome Extensions”, Academy of Management Review, 25 (1): 10–11.

Payne, M. and Ballantyne, D. (1993) Relationship Marketing: Bringing Quality,Customer Service and Marketing Together, Oxford: Butterworth-Heinemann.

Pheysey, D. (1993) Organisational Cultures: Types and Transformations, London:Routledge.

Piercy, N. and Cravens, D. (1995) “The Network Paradigm and the MarketingOrganization: Developing a New Management Agenda”, European Journal ofMarketing, 29 (3): 7–34.

Powell, W. and DiMaggio, P. (1991) The New Institutionalism in Organizational Analysis,Chicago, Ill.: University of Chicago Press.

Ravald, A. and Groonroos, C. (1996) “The Value Concept and RelationshipMarketing”, European Journal of Marketing, 30 (2): 19–31.

Ring, P. and Van de Ven, A. (1992) “Structuring Cooperative Relationships betweenOrganizations”, Strategic Management Journal, 13 (7): 483–98.

Sarkar, M., Echambadi, R., Tamer Cavusgil, S. and Aulakh, P. (2001) “The Influ-ence of Complementarity, Compatibility and Relationship Capital on AlliancePerformance”, Journal of the Academy of Marketing Science, 29 (4): 358–73.

Sawhney, M. and Zabin, J. (2001) Seven Steps to Nirvana: Strategic Insight intoEbusiness Transformation, New York: McGraw-Hill.

Schwartz, R. (2005) “The Contracting Quandary: Managing Local Authority–VNPO Relations”, Local Government Studies, 31 (1): 69–83.

Scrivens, E. (1991) “Is There a Role for Marketing in the Public Sector?”, PublicMoney and Management, Summer: 17–23.

Selnes, F. (1998) “Antecedents and Consequences of Trust and Satisfaction inBuyer–seller Relationships”, European Journal of Marketing, 32 (3–4): 305–22.

Sheaff, R. (1991) Marketing for Health Services: A Framework for Communications,Evaluation and Total Quality Management, Oxford: PsychoBabel.

Sharma, N. and Patterson, P. (1999) “The Impact of Communication Effectivenessand Service Quality on Relationship Commitment in Consumer, Professional Services”, Journal of Services Marketing, 13 (2): 150–70.

Sheth, J. and Parvatiyar, A. (2000) Handbook of Relationship Marketing, ThousandOaks, Calif.: Sage.

Sheth, J., Sisodia, R. and Sharma, A. (2000) “The Antecedents and Consequencesof Customer-centric Marketing”, Journal of the Academy of Marketing Science, 28 (1): 55–66.

Tsai, W. (2000) “Social Capital, Strategic Relatedness and the Formation ofIntraorganizational Linkages”, Strategic Management Journal, 21 (9): 925–39.

Veloutsou, C., Saren, M. and Tzokas, N. (2002) “Relationship Marketing: What if. . . ?”, European Journal of Marketing, 36 (4): 433–49.

Vickers, J. and Yarrow, G. (1988) Privatization: An Economic Analysis, Cambridge,Mass.: MIT Press.

Walsh, K. (1991) “Citizens and Consumers: Marketing and Public Sector Manage-ment”, Public Money and Management, June–August: 9–16.

Walsh, K. (1994) “Marketing and Public Sector Management”, European Journal ofMarketing, 28 (3): 63–71.

Waterman, R. and Meier, K. (1998) “Principal–agent Models: An Expansion?”, Journalof Public Administration Research and Theory, 8 (2): 173–202.

Williamson, O. (1985) The Economic Institutions of Capitalism: Firms, Markets,Relational Contracting, New York: Free Press.

Wistow, G., Knapp, M., Hardy, B. and Allen, C. (1994) Social Care in a MixedEconomy, Buckingham: Open University Press.

Wright, G. and Taylor, A. (2005) “Strategic Marketing in Health Care”, PublicManagement Review, 7 (2): 203–24.

LEADING ACROSS FRONTIERS:HOW VISIONARY LEADERS

INTEGRATE PEOPLE, PROCESSES,STRUCTURES AND RESOURCES

Barbara C. Crosby, John M. Bryson and Melissa M. Stone

Within collaborations aimed at solving complex public problems, leadershipis clearly important in achieving successful outcomes (Bryson, Crosby andStone 2006; Ansell and Gash 2008). We are especially interested in betterunderstanding how leaders work across sectoral, organizational and culturalboundaries to bring diverse groups of people together to work out sustain-able remedies for such problems. We see the leadership practice of these leaders as integrative – that is, they help organizations integrate people, processes, structures and resources in semi-permanent ways. Integration issimilar to collaboration, but we think the former term captures more directlythe need for leaders and constituents to move back and forth across bound-aries and build linking pathways and other commonalities.

This chapter explores the visionary practices of integrative leaders in a collaborative transportation initiative in the Minneapolis-St Paul regionof the United States (Bryson, Crosby and Stone 2008). The focus is on visionary leadership practices because they are vital to helping diverse stake-holders develop a shared understanding of a public problem and potentialsolutions, as well as commit to work for new policy regimes (Crosby andBryson 2005a). The chapter weaves together three theoretical strands – leadership, collaboration and cross-boundary work – to supply the basic fabric of our analysis. To begin, we explain our research design – includingthe theoretical strands, definitions and methods. The following section describes the Urban Partnership program developed in recent years by theUS Department of Transportation (USDOT), and details the successfuleffort of a group of Minnesota transportation officials, local governmentsand policy advocates to secure one of the major grants (Urban Partner-ship Agreements or UPAs) awarded through the program. We then offerfindings about how visionary leaders helped diverse stakeholders develop shared

understandings and commitments that allowed the group successfully to com-pete for and implement an Urban Partnership project. Finally, we presentlessons and conclusions.

Research design: theory and methodology

In order to understand how leaders exercised visionary leadership in this case, we draw theoretical strands from several bodies of work: Crosby andBryson’s work on how leaders inspire and mobilize diverse constituents to tackle complex public problems in shared-power environments where no one is wholly in charge (Crosby and Bryson 2005a, 2005b); Huxham and Vangen (2005) and their colleagues’ studies of collaboration; and ourown review of the literature on cross-sector collaboration (Bryson, Crosbyand Stone 2006). We also consider the role of boundary objects, boundaryexperiences and boundary groups (Carlile 2002, 2004; Feldman et al. 2006;Kellogg, Orlikowski and Yates 2006; and Bryson, Crosby and Bryson 2009),and the role of technology (Orlikowski 2000) in facilitating the creation and maintenance of multi-stakeholder, multi-sector collaborations. (Theattention to technology emerged from our study of the Urban Partnershipcase, because computer-assisted “intelligent transportation systems” were a key component.)

Our initial research on this case was aimed at testing and refining our cross-sector collaboration framework (Bryson, Crosby and Stone 2006) andunderstanding more concretely the factors that affect the success of cross-sector collaborations. The results of that research are reported in “Colla-boration in Fighting Traffic Congestion: A Study of Minnesota’s UrbanPartnership Agreement” (Bryson, Crosby and Stone 2008). We explain ourcase-development methods after an in-depth consideration of our theoret-ical fabric.

Theoretical strands

Crosby and Bryson’s Leadership for the Common Good framework is the organ-izing strand for this study. Strands focusing on cross-sector collaborationand cross-boundary work are woven in as we seek to illumine leadership aspectsof this particular case of collaborative public problem-solving.

Leadership for the common good

Crosby and Bryson (2005) emphasize the importance of, and intercon-nections between, visionary, political and ethical leadership practices for helping diverse stakeholders make sense of a public problem, commit to doing something about it, obtain needed policy and implementation decisions,and enforce new shared-power arrangements. These practices are linked

LEADING ACROSS FRONTIERS

together in a multi-dimensional view of power (Figure 12.1) that highlightsthe creation and communication of meaning in formal and informal forums(the main practice of visionary leadership), policy-making and implementa-tion in formal and informal arenas (the main practice of political leadership),and conflict resolution and normative regulation in formal and informal courts (the main practice of ethical leadership). Each of the practices has threedimensions: (1) observable action; (2) the rules, modes, media, methods thatunderpin action; and (3) the taken-for-granted, deep social structures of language and meaning (Crosby and Bryson 2005a; Giddens 1979, 1984).

We argue that, in order to operate effectively in environments in which thepower to resolve a public problem is dispersed among many individuals, groupsand organizations, leaders should direct their attention and energies to the middle dimension – that is, the rules, media and modes that draw on thedeep structures in the third dimension and shape observable action in thefirst dimension. An especially important element of the middle dimension is the design of formal and informal forums, arenas and courts, the socialsettings in which observable action occurs. We argue that, in pursuing thecommon good, visionary leaders emphasize the design and use of forums,political leaders emphasize the design and use of arenas, and ethical leadersemphasize the design and use of courts.

Ideas, Rules,Modes, Mediaand Methods

2nd dimensionof power

Human Action

1st dimensionof power

Deep Structure

3rd dimensionof power

Policy making and implementation capabilitiesDomainAgendaPlanning, budgeting, decision making, and implementation methodsAccess rules

Policymaking and

implementation

Principlesof

domination

Social, Political, Economic, and Natural Environments

Communicative capabilityInterpretive schemesRelevanceNorms of pragmatic communicationModes of argumentAccess rules

Creation andcommunication

of meaning

The Use of Forums The Use of Arenas The Use of Courts

Principlesof

signification

Conflict management and sanctioning capabilitiesNormsJurisdictionConflict management methodsAccess rules

Managementof residual conflict

and enforcement ofunderlying norms

Principlesof

legitimation

Figure 12.1 The triple three-dimensional view of power. From Crosby, B. andBryson, J. (2005) Leadership for the Common Good, San Francisco:Jossey-Bass, p. 409

Thus, visionary leaders who wish to develop common understanding ormeaning around traffic congestion would focus on the rules, media and modesthat shape communication in formal and informal forums. Perhaps they decideto hold a workshop highlighting different perspectives on traffic congestion.They will establish rules to ensure that some people are speaking during the debate and others are not. They will structure the workshop agenda and perhaps require that ideas be presented in a certain format. They willdevelop an invitation list and strategies for ensuring that certain people areor are not in the audience. They will use methods that winnow and synthe-size disparate ideas.

Cross-sector collaboration

In addition to arguing for the necessity of visionary, political and ethical leadership for remedying complex public problems, we argue that remedy-ing complex public problems is likely to involve cross-sector collaborationof some sort (Bryson, Crosby and Stone 2006), and thus leaders will needan understanding of what fosters successful cross-sector collaboration. Suchcollaborations are unlikely to get off the ground unless potential partnershave some agreement on a need or opportunity that requires collaboration(Mattessich, Murray-Close and Monsey 2001). Collaborations are unlikelyto get very far after that unless partners develop enough trust to continueworking together, find ways to deal with power imbalances, cope with shifting membership, develop supportive processes and structures, and havea system of evaluating outcomes and fostering accountability (Winer and Ray1994; Huxham and Vangen 2005; Bryson, Crosby and Stone 2006).

Our analysis of the literature on cross-sector collaborations indicated that initial conditions, process and structure, contingencies (such as power)and accountability mechanisms all have important effects on the formation,operation and outcomes of these initiatives (Bryson, Crosby and Stone 2006)(see Figure 12.2). Initial conditions that seemed to contribute to the forma-tion of collaborations included turbulence in the environment, history of sector failure, supportive politics, pre-existing formal and informal networks,and general agreement on a problem or opportunity that would be the focusof the collaboration. Crucial process components seemed to be methods ofbuilding and exercising leadership, creation of various initial agreements amongcollaborating partners, development of the collaboration’s legitimacy, cre-ation and maintenance of trust, effective conflict management, and planning.Crucial structural components were governance arrangements, membershipcriteria and characteristics, and blend of hierarchy and network. Import-ant contingencies included power imbalances and competing institutional logics. We concluded that a measure of cross-sector collaborations’ successis the extent to which they produce “public value”, a term coined by MarkMoore (1995) and which we define as fulfilling public purposes, mandates,

GENERAL ENVIRONMENT• Turbulence• Competitive and institutional elements

SECTOR FAILURE

DIRECT ANTECEDENTS• Conveners• General agreement on the problem• Existing relationships or networks

OUTCOMES• Public value• First-, second-, and third-order effects• Resilience and reassessment

ACCOUNTABILITIES• Inputs, processes, and outputs• Results management system• Relationships with political and professional constituencies

• Building and exercising leadership• Forging agreements• Building legitimacy• Building trust• Managing conflict• Planning

BOUNDARYEXPERIENCES

(especially forums),OBJECTS (incl. visions),

AND GROUPS

• Type of collaboration• Power imbalances• Competing institutional logics

Formal and informal• Membership• Structural configuration• Governance structure

INITIAL CONDITIONS

CONTINGENCIESAND CONSTRAINTS

OUTCOMES ANDACCOUNTABILITIES

ACTION/PROCESS STRUCTURE ANDGOVERNANCEFormal and informal

Figure 12.2 A framework for understanding leadership in cross-sector collaborations.

organizational missions and the expectations of the citizenry at reasonablecost. To construct a reliable picture of collaborations’ outcomes, participantsand analysts will need to rely on an accountability system that tracks inputs,processes and outcomes; uses a variety of methods for gathering, inter-preting, and using data; and relies on strong relationships with key politicaland professional constituencies.

This chapter focuses on the process components, but also on their inter-action with initial conditions, structure, contingencies and outcomes. Our questions specifically are: How do visionary leaders help collaborating partners from different sectors develop enough of a shared purpose to formthe basis of initial agreements? How do they use planning processes to buildshared understandings and commitments to craft and implement solutions?How do they use the design of settings (with their partially visible rules) tocreate a shared purpose and set of commitments, and to build trust, estab-lish legitimacy and manage conflict? We think attention to the intermediatelevel of power – rules, modes, media and methods – can be very helpful inanswering these questions, but we shall also use insights from the literaturethat emphasize the important role of boundary objects, boundary experiencesand boundary groups in helping people connect with perspectives very different from their own and commit to a new, shared perspective. In thiscase, technology also appears to operate at the intermediate level of poweras an integrative underpinning of observable action.

Boundary experiences, objects and groups

Boundary experiences are “shared or joint activities that create a sense ofcommunity and an ability to transcend boundaries among participants”(Feldman et al. 2006: 94; Feldman and Khademian 2007). UPA was forgedout of the self-conscious design and use of such experiences; it continues, atleast in part, because of them as well. Boundary objects are “physicalobjects that enable people to understand other perspectives” (Feldman et al.2006: 95). Beyond that, boundary objects can facilitate the transformationof diverse views into shared knowledge and understanding (Carlile 2002;Kellogg, Orlikowski and Yates 2006). Attention to boundary objects first developed in a production or manufacturing context, where the objectsoften were scale models or mockups, but since then the idea has beenextended to include less material objects. Boundary objects and their devel-opment help participants make sense of their world, what they may want to do with it, and why; and, in doing so, the exercise helps participants connect people, ideas and other kinds of actors into a way forward. Saiddifferently, boundary objects act as a kind of “transitional object” (Winnicott1953) or “facilitative device” (de Geus 1988) from here-and-now real possi-bilities to the there-and-then actualities. Boundary objects may be used todeal with what Carlile (2004) calls “syntactical” problems, or incompatible

codes, routines or protocols; “semantic” problems, resulting from differencesin meaning, assumptions or context; or “pragmatic” problems, caused by dif-fering stakeholder interests that can be resolved only if people alter some oftheir existing knowledge or ways of doing things.

Boundary organizations or boundary groups are “collections of actors whoare drawn together from different ways of knowing or bases of experiencefor the purpose of coproducing boundary actions” (Feldman et al. 2006: 95).Examples include cross-boundary networks, taskforces and teams; coordinat-ing committees; and representative policy-making bodies. Minnesota’s UPAis perhaps best-understood as a boundary organization; it is mostly a “vir-tual organization” comprised of many members, including the MinnesotaDepartment of Transportation (MnDOT), Metro Transit (the regional trans-portation authority) and the City of Minneapolis, and a host of other organizations, groups and individuals participate.

Other theoretical threads

Theories about cross-cultural communication also may be important in this case,since cultures at the state and local level may be quite different from the cul-ture of federal agencies. (Culture here refers to habitual ways of doing things,distinctive organizational and institutional rituals, core values, and routinesand logics [Schein 2004].) The cultures of government agencies are likely tobe quite different from those of nonprofits and business. A similar concept,“competing institutional logics”, appears in organizational theory and neo-institutional views of the environment to emphasize the deep-seated and oftencontradictory norms and views of action embedded in the logics of bureau-cracy, democracy, markets, and so forth (Friedland and Alford 1991).

Studies of collaboration have paid little attention to the role of com-puterized information technology in facilitating or hampering cross-sectorcollaboration. Orlikowski (2000) has offered a “practice lens” for studyingtechnology in organizations, and we draw on her insights. Our analysis ofthe Minnesota UPA revealed that technology played several important roles:as its own driving force, as a solution, and as a motivator, facilitator andpositive political force (Bryson, Crosby and Stone 2008).

Case-study methodology

Our extended study of the Minnesota UPA project used a simplified versionof the framework presented in Figure 12.2 to develop the questions we posedto those closely involved in advocacy, conceptualization and management ofthe implementation of the project. We conducted semi-structured interviewswith twenty-six individuals involved in Minnesota’s UPA. In selecting oursample, we paid careful attention to gaining perspectives from individualsat multiple levels of government and with varying levels of responsibility

and authority over the UPA implementation, including federal officials, state legislators, MnDOT and Metro Transit staff, policy advocates, legislators,and members of local transportation groups.

The interview protocol included questions related to the interviewee’s background in the transportation field, the initial conditions leading up to the UPA collaboration, the structures and processes of decision-making,and the outcomes and accountability processes involved. A note-taker accom-panied the interviewer to record the interviewee’s comments verbatim on alaptop computer, and these notes were then imported into the qualitativeanalysis software program QSR NVIVO.

Important to our design and analysis was feedback from key practitionersin the local transportation field. We convened an Advisory Group comprisedof leaders from each of the primary UPA partnership organizations to pro-vide feedback on our initial findings and draft reports. Archival newspaperarticles and other publications formed the basis of a secondary data-collection effort focused on capturing the story of UPA development andimplementation as reported by local newspapers and publications in the recipient states, with particular emphasis on the legislative and political processes necessary for successful implementation of the UPA policy.

We developed a thematic coding structure based on our original cross-sector collaboration paper, input from the Advisory Group, and discussionamong research-team members. The software also allowed us to disaggre-gate thematic results into categories based on the characteristics of interviewparticipants, such as their work affiliation and job title.

Background of the Urban Partnership Agreement

Traffic congestion in US metropolitan areas has become an exceedingly com-plex, or “wicked”, problem in the last four decades. The habitual approachto solving the problem – building more freeways – increasingly ran into limitations in the form of citizen protest and competing demands on publicbudgets. Plus, when new or expanded freeways or other roads were built theyoften did not reduce congestion in the long run. Soon the new lanes werealso clogged with traffic as more people moved into the areas and low-density development continued to spread outward from city centers. By the1960s transportation analysts with economics training were beginning to arguethat the only way actually to reduce traffic congestion was to price the useof the roads during peak hours of use. Referring to the laws of supply anddemand, they argued that demand would go down if drivers had to pay foraccess. The analysts argued that some drivers would pay the fee, but otherswould take alternative, uncongested routes, vary their driving time, take the bus, or stay at home. By the 1990s, policy entrepreneurs were imagining integrated transportation systems that relied on congestion pricing, transit,a variety of advanced technologies, and telecommuting.

From the 1960s through the 1990s, the idea of using pricing to managetraffic congestion had difficulty getting off the ground. Feasibility studies anda few pilot projects were tried, but elected officials and citizens generally werenot convinced that the approach would work. Citizens also objected to pay-ing a fee for facilities they felt they had already funded through their taxes.By the late 1990s, however, congestion was getting even worse in many urbanareas, and a greater number of public officials were realizing that they couldnot build their way out of the problem.

Within USDOT during the George W. Bush Administration, Tyler Duvall,assistant secretary for transportation policy, began working with a few othertop transportation officials to move from researching congestion pricing to mounting larger-scale demonstrations. One of his key allies was Mary Peters,the then administrator of the Federal Highway Administration, but there were a number of others in the department, including Patrick DeCorla-Souza, who was a long-time advocate of congestion pricing. Duvall tried to con-vince USDOT Secretary Norm Mineta to make congestion pricing a federalpriority. Initially, Mineta was skeptical but after a top-level strategy meet-ing in 2006 agreed to make the shift, and congestion pricing was includedin the department’s 2006 Strategy Statement. Duvall, Peters, David Horner(the chief counsel of the Federal Transit Authority) and others then begandesigning a demonstration project to channel funding to major metropolitanareas that would tackle congestion with a set of complementary strategiescalled “the four Ts”: transit, technology, tolling and telecommuting. The designers thought that integrating the four strategies would provide the big-gest payoff in terms of congestion pricing. They were able to secure about$120 million in departmental discretionary funds to put into what becameknown as the Urban Partnership project, which was designed to demonstratewhether congestion pricing had a clear positive impact. The plan was to create a competitive request for proposal (RFP) process that would resultin approval of a few highly promising projects.

Soon, however, a much larger amount of money became available whenCongress suspended its usual practice of allowing members to earmarktransportation funds. As a result, the pot for the project eventually grew to $1.1 billion. In addition, Mary Peters became US Secretary of Trans-portation, allowing her to champion the program from the top position inthe department.

In Minnesota, congestion pricing advocates, state and local officials, andtransit supporters began discussing participation in the Urban Partner-ship program, officially announced at the end of 2006. Minnesota, after all,was the site of one of the country’s most successful congestion-pricingexperiments, in the form of the MnPass project on highway I-394 – anIntelligent Transportation System (ITS) application of dynamic pricing to a segment of I-394 in the western part of the Minneapolis-St Paul metro-politan region.

Minnesota Department of Transportation (MnDOT) officials decided, after some initial reluctance, to submit a proposal for an Urban Partnershipgrant in collaboration with the Metropolitan Council, which operates thebus transit system (Metro Transit) for the Minneapolis-St Paul region. Soonafter, the Citizens League (a nonprofit public-policy study group focusingon Minneapolis-St Paul), the University of Minnesota’s Center for Trans-portation Studies, and the State and Local Policy Program at the university’sHumphrey Institute featured the Urban Partnership program at their RoadPricing Summit on 1 February 2007. It was here that Rick Arnebeck fromMnDOT announced that the department would seek a UPA grant. Tyler Duvallalso spoke at the summit.

MnDOT project leaders assembled an interagency Steering Committee tooversee the proposal development process. In addition to individuals fromMnDOT and the Metropolitan Council, the committee over time grew toinclude local officials from highly congested traffic corridors, county officials,and University of Minnesota experts. MnDOT hired SRF Consulting Groupto prepare the actual grant proposal. John Doan of SRF played a key role in the drafting process; he was a former MnDOT employee who hadworked on congestion pricing while there.

Since the proposal was due at the end of April, the Steering Committeemembers knew that they had to obtain agreement among numerous stateand local parties about the main components of the proposal. For example:In which locations would congestion pricing be applied? What form wouldit take? What would be the implications for bus services and routing? Whattechnological innovations would be emphasized? What role would tele-commuting play?

The committee organized a half-day workshop in March and several sub-sequent meetings to help numerous stakeholders consider possible answersto these questions and develop a consensus about what should be includedin the proposal. Additionally, project supporters worked behind the scenesto make sure that powerful legislators, the governor and the lieutenant-governor would support the form of tolling that would be included in theMinnesota UPA proposal.

At times, project advocates worried that disagreements about proposal components would sink the effort, but eventually the Steering Committeeand outside advocates obtained enough consensus and compromise to be able to submit a strong proposal, focusing on the I-35W corridor south ofMinneapolis and on downtown Minneapolis bus routes that linked to thecorridor. Minnesota’s proposal was selected as one of the nine semi-finalistsannounced by USDOT in June. The semi-finalists then were invited to pre-sent their plans to USDOT, and in August the nine were winnowed to fivefinalists – Minneapolis, Seattle, New York, San Francisco and Miami. (Inthe spring of 2008, New York would drop out, and Los Angeles and Chicagowould be added.)

The total UPA grant to Minnesota was $133.3 million to be matched with$55.2 million in funds from the state legislature and Metropolitan Council.In addition to approving the match, state legislators would also have to approvetolling authority for the I-35W corridor. The UPA partners had approximatelyone year to assemble all components of the implementation plan.

Once Minnesota was chosen as a finalist, the UPA Steering Committeewent into implementation mode. It became a smaller, more operations-oriented group, and MnDOT put Nick Thompson, operations manager, incharge of day-to-day oversight of the operational aspects of the project. Atthe same time, the Metropolitan Council transit officials and local govern-ment partners began working on their pieces of the project, while legislatorsand MnDOT senior officials worked on legislative strategy. Ultimately, the 2008 session of the Minnesota Legislature approved the required statematching funds as well as needed MnDOT authority for implementing the UPA.

How sponsors and champions exercised visionary leadership in the UPA case

Two main types of leaders – sponsors and champions – seem necessary forcarrying out successful collaborative efforts to tackle public problems (Crosbyand Bryson 2005a). Sponsors bring formal authority, financial resources andlegitimacy to the endeavors. Champions typically bring tireless commitment,networking skills and often significant informal authority. They are willingto take risks in the service of potential payoff. They understand the policy-change process and take a long view. In this case, federal-level championsand sponsors worked together to construct the Urban Partnership program,and state and local champions and sponsors put together Minnesota’s successful campaign to obtain and implement a UPA. Among the federalleaders, Mary Peters was initially a champion and became a sponsor. TylerDuvall, Patrick DeCorla-Souza and David Horner were champions. InMinnesota, champions included Lee Munnich, Adeel Lari, Bob DeBoer, members of the I-35W Solutions Alliance (consisting of local officials in the 35W corridor) and a MnDOT middle manager. Senior executives atMnDOT and Metro Transit were important sponsors.

Our analysis focuses on how these champions and sponsors exercised vision-ary leadership in a way that integrated stakeholders, processes, structuresand resources from different sectors and governmental levels to tackle a shared problem. As noted earlier, visionary leaders do the crucial work ofhelping constituents make sense of a public problem and commit to doingsomething about it. As these leaders assemble collaborations, they focus on developing agreement on the collaboration’s purpose or mission, and securing commitment to producing and implementing strategies for carry-ing it out. In carrying out this work, visionary leaders emphasize the social

practice of creating and communicating shared meaning in formal andinformal forums.

Visionary leadership comprises three main tasks: interpreting the need or opportunity and giving direction about how to respond to it, offering compelling visions of the future, and adeptly designing and using formal andinformal forums. Let us now consider the processes by which sponsors and champions in this case carried out those tasks in ways that integrateddifferent government units and levels and government and nongovernmentpartners.

Seizing opportunities to provide interpretation and direction

Visionary leaders help make public problems real through a problem-definition process: they make social needs and opportunities visible, detailtheir causes and consequences, and frame them in ways that appeal to diversestakeholders. They also engage in solution-seeking processes, chiefly elicitingand championing new ideas for dealing with the problems.

Making societal needs or opportunities visible

In this case, existence of the problem of urban traffic congestion was fairlywidely known. Highway-users experienced the problem daily, governmentofficials fielded complaints about it, academic and government reports detailedits pervasiveness and costs, and public-opinion surveys often put it high onthe list of citizen concerns (Metropolitan Council 2007).

Detailing causes and consequences

Citizens and analysts alike tend to agree that a major factor in urban traffic congestion is too much reliance on cars and (possibly) highways andland-use patterns, especially urban sprawl. This discussion has been goingon for decades.

In the interviews conducted for our UPA study, analysts at the federallevel also pointed to the existence of federal transportation programs (andfunding) in separate silos, marked “highways”, “buses”, “intelligent trans-portation systems” and “non-motorized transport”. Analysts also realizedthat these arrangements provided little incentive for state and local trans-portation agencies and decision-makers to coordinate their work. Compre-hensive and coordinated planning to ease a multi-faceted problem like congestion was difficult indeed. A more behind-the-scenes concern withinUSDOT was that federal highway programs in particular were not focusedand data-driven.

The at-least-partially-shared problem diagnosis lowered barriers amongprospective UPA partners in Minnesota. All agreed that urban congestion

was bad and getting worse. They did focus on different parts of the prob-lem: legislators tended to see the congestion located in their districts; the I-35W Solutions Alliance was fighting against congestion in a particular highway corridor. The Minnesota partners had enough agreement on the problem and the shared mission of remedying it that they could worktogether, in keeping with Huxham’s finding that collaborations can proceedwithout complete agreement on aims (Huxham 2003).

Framing the problem and solutions

Stakeholders, including the general public, generally agreed that traffic congestion was a public problem and that therefore government has someresponsibility for doing something about it. Government reports, citizen surveys, media accounts and scholarly reports portrayed traffic congestionas a public bad because of widespread and multiple harmful effects: pollu-tion, personal stress, economic toll. An example would be the Critical Issuesin Transportation report published by the Transportation Research Board(2007), a division of the nonprofit National Research Council.

As noted above, federal transportation analysts and other anti-congestioncampaigners had begun to argue that congestion was a complex problemthat required comprehensive integrated solutions. A fragmented approach was characterized as failure-prone, and an integrated approach was deemedpotentially very effective. The integrative frame was expressed in the linkingof the “4 Ts” (tolling, transit, technology and telecommuting) in the grantapplication process and public proclamations about it.

Money also helped convey the idea of an integrated system across bound-aries. In the Road Pricing Summit sponsored by the Citizens League and infollow-up informal conversations, people who were not necessarily attractedto pricing or the 4-T idea did respond to the lure of major federal funding.The funding conveyed a sense of legitimacy (federal officials are willing toput big money behind this idea) and it also promised significant impact. Thispromise helped motivate potential partners to begin thinking how to aligntheir goals and favored projects with what federal officials were proposing.It also helped overcome resistance from transportation-planners concernedabout the possible need to rethink projects already underway. Additionally,Minnesota partners needed incentives to accept the risk that they would sinktime and effort into something (the UPA grant application) that might notsucceed. In other words, the large size of the grant was a powerful induce-ment (Stone 2002). Perhaps the use of the word partnership had positive impactin that it conveyed a sense that federal officials were willing to let state andlocal applicants come up with program specifics and that they would stayinvolved as the projects went forward.

Additionally, what might be called a transportation corridor frame had become a familiar way of thinking about transportation projects –

transportation planners and policy-makers had begun to think beyondrebuilding a highway or establishing a bus route, to considering the waysvarious modes of travel operated together in a corridor that included at least one major traffic route and feeder routes. The corridor idea helped linktogether modes and stakeholders along the corridor.

The prospect of integrating advanced technology throughout the projecthelped stakeholders see the UPA as an exciting opportunity to innovate. As one interviewee stated, people are attracted by the possibility of beinginnovators, because “[i]t’s exciting to implement new technology”. Anotheradded that an incentive for MnDOT was “another opportunity to be cutting edge”. The UPA was therefore a technology-assisted motivating orattractor force (Allen and Cherrey 2000).

A sticking point was the framing of tolling as a “public bad”. The veryword “tolling” has a negative connotation for many citizens. They object topaying tolls for something they see as a public good. They argue: “I’ve paidfor the highway with my taxes. Why should I pay again?” The supportersof “tolling” or “pricing” in this case tried to replace this negative frame with a positive “public good” frame. Tax dollars, they suggested, simply guarantee that highways are universally available on demand as long as theircapacity is not exceeded. After that users might legitimately pay a price foraccess to free-flowing lanes. Moreover, thanks to transponder technology,that price can be carefully calibrated to respond to high demand or peakperiods. In effect, free flow becomes worth more when more people want it. The label “dynamic pricing” captured this argument.

Pricing advocates still had to overcome the perception that charging for the use of one or more lanes was taking those lanes away or subtract-ing capacity from the system. They countered that pricing actually increasedthe numbers of cars that a highway can accommodate. This argument is some-what less straightforward than the argument for pricing. One advocate usedthe analogy of pouring rice into an hourglass: that is, if many grains of riceare poured rapidly into an hourglass they will soon overwhelm the capacityof the narrow neck, and almost no grains will get through after a certainpoint in time. Yet if the rate of entry is slowed enough, which presumablypricing would do, all the grains will get through. Advocates quite literallydid this; in other words, they used the rice and hourglass as a boundary objectto facilitate shared knowledge and understanding. Demonstration projects– such as MnPass on I-394, and the Stockholm and London experiences –that show the beneficial effects of pricing were also crucial for making thisless-than-intuitive argument stick.

Championing new ideas for dealing with the problems

Congestion pricing was a relatively new idea for some stakeholders. The effec-tiveness of its use outside the US and in a few places inside the US helped

champions make their case to different stakeholder groups. Effectiveness or efficiency arguments appeal across sectors and government levels. The idea of integration seemed to help as well. Supporters of telecommuting, busrapid transit, congestion pricing and “intelligent systems” could all see howtheir favored part of the congestion solution could be advanced but alsoenhanced by being meshed with other solutions within the UPA program.Making funding contingent on an integrated UPA grant application was ameans that federal transportation officials used to ensure that this idea wouldhave traction. The federal designers of the UPA program also built in fund-ing and mandates for evaluation so that the program appealed to rationalplanning advocates.

Federal and state champions had gotten their tolling idea fully developedin preparation for the policy window (Kingdon 1995) that opened when trafficcongestion emerged as a priority transportation-policy focus in the final two years of the Bush administration, and the transportation secretary pickedup substantial discretionary funding because of the impasse over earmark-ing. They instituted a competitive process that required complementary, silo-busting strategies and “forced creative thinking”, as one interviewee said.They even took a bit of comfort from any resistance they encountered – theyinterpreted it as a sign of innovativeness. Meanwhile, the RFP, competitivegrant process was a well-known routine for most players. So it fitted withintheir normal frames of reference.

Perhaps in their enthusiasm for alliteration, the federal policy entre-preneurs slipped up. In Minnesota, tolling raised the hackles of the governor,many legislators, and citizens. Paying to use a road already paid for by taxesseemed anathema to them. Indeed, any tolling on Minnesota highways hadto be expressly permitted by the state legislature. Arguments that conges-tion pricing added capacity were not enough to win the governor’s support.To win him over, the assemblers of Minnesota’s UPA application had toapply dynamic pricing to a shoulder lane, so that existing lanes were not “taken away” by tolling. Now the shoulder lane idea is being considered in other states. Meanwhile, USDOT decision-makers had to be willing tomove away from their desire to price existing lanes, if they were to approvethe Minnesota application. They ultimately selected Minnesota as a UPAfinalist because transportation agencies in the state had a reputation for competence, plus success with congestion pricing, and thus federal officialsviewed them as legitimate and trustworthy partners.

Technological advances made “dynamic pricing” possible as a solution to traffic congestion and allowed motorists to use transponders rather thantoll booths to pay. According to one interviewee, technology allows “roadpricing without significant transaction costs”. He also noted that technologyimproves transit services because it allows “buses to travel with shorter headways” and makes transit service more predictable and reliable by, forexample, providing “real-time” traveler information. All of this enhances the

attractiveness of Bus Rapid Transit (BRT) and “thereby creates a virtuouscycle for transit – the more appeal, the more demand for it, the greater thefrequency with which it’s provided, the better the economies of scale, thelower the cost per traveler [and therefore the more demand for it]”.

Offering compelling visions of the future

Sponsors and champions engage in translation and construction processesas they attempt to make an alternative future “real” in the minds of con-stituents. To be persuasive, they must translate abstract ideas like congestionpricing into accessible and appealing words and imagery that can provide aclear pathway from current conditions to a desirable future. Elements of thepath and the future state must be constructed or woven together in a plaus-ible and compelling way. In this case, the vision communicated explicitly and implicitly by sponsors and champions of the UPA program includedsmooth-flowing highways on which substantial numbers of vehicles usetransponders to take advantage of priced lanes, technologically sophisticatedside-boarding buses operate a lot like light rail trains, commuters con-tinually are fed up-to-date information about travel times, park-and-ride facilities are available, new commercial and residential development springsup around bus hubs, and more people use the Internet to work from homeor hook up to carpools. The vision was communicated through verbalimagery, cross-boundary objects (like the RFP, drawings, maps and reports)and groups (like Minnesota’s UPA Steering Committee) that made it potentially real and thus believable. This may be where trust comes in – perhaps visionary leaders foster shared interpretation of others’ motives, competencies and commitments.

Designing and using formal and informal forums

Structure and process come together in the design and use of formal andinformal forums, the settings in which leaders and constituents engage in thedialogue, discussion and debate that may result in shared understandings of public problems and solutions – in this case, urban traffic congestion andintegrated arrangements of transit, tolling, technology and telecommuting.UPA sponsors and champions put together a number of cross-boundaryforums at the national, state, regional and local levels. For example, in January2006 about eighty leaders of transportation agencies that comprise USDOTparticipated in an offsite planning retreat, where they identified metro-politan highway congestion as the nation’s biggest transportation problemdeserving priority. An urban congestion working group was formed to comeup with an action plan in sixty days. In these high-level forums, an eco-nomics/pricing frame was defensible because congestion-pricing initiativeshad succeeded in the US and abroad and because other solutions clearly were

not having much effect. What emerged from these forums was the vision of an integrated system, branded with the “4 Ts” label. The “4 T” label wasitself a kind of boundary object, or at least a summary term that carried somewhat shared meaning across audiences.

Once the UPA program was defined, federal and local champions usedfollow-up forums to alert potential applicants to the program’s existence and its potential to provide significant remedies for the problem. Bridgingacross the boundaries between national and local was facilitated by the existence of individuals and groups in the Minneapolis-St Paul area who hadparticipated in national transportation forums, supported and publicizedMinnesota’s successful MnPass program, and shared the economics pricingframe. Some had worked together on corridor projects. Some MTA andMnDOT employees had collaborated as partners in a group called TeamTransit. Several interviewees pointed to the pre-existing relationships amongstakeholders as being a key contributor to their willingness to get involvedin shaping the UPA. This finding reinforced our previous study (Bryson, Crosbyand Stone 2006) indicating that pre-existing networks and relationshipswere an important initial condition affecting the formation of cross-sectorcollaborations.

Stakeholder forums convened during Minnesota’s application processincluded disparate groups, and facilitators used methods designed to fosterinclusion and balance power. The I-35W Solutions Alliance was powerful inthese settings because members had their act together. A sense of urgencywas supplied by the short timelines that USDOT had imposed for sub-mitting Urban Partnership applications. Agreements from these forums andSteering Committee meetings solidified into the actual grant application.

Forums may be thought of as structures that mediate between deep socialstructures such as language and observable action such as the debate anddiscussion that occurs in a workshop. Forums convened as part of a cross-sector collaboration are shaped by and also facilitate collaborative processesto carry out the work of the collaboration. They can be boundary experi-ences, produce boundary objects and often result in formation of boundarygroups, such as taskforces and steering committees. Recalling the syntactical,semantic and pragmatic functions of boundary objects, we conclude that syntax was not much of an issue in this case, since most of the partners, regardless of their government level or sector, spoke “transportation-ese”.They did have semantic differences – for example, over tolling or intermodalapproaches. They had to work out what integration across modes meant.They had to grapple with how much change would be needed if differentprofessional groups associated with different modes had to work together.Pragmatics would require building a coalition, and boundary objects couldbe something around which people could coalesce and make their ownmeaning and align their interests. The proposal that was submitted and woncontained all of these elements. Supporters quite literally had to agree on

the syntax in the proposal; they had to agree on what the words meant; andthe proposal itself represented a shared agreement or treaty that reconciledinterests and committed the signers to alter some of what they knew andhow they would do things.

Another important boundary object was the USDOT request for proposal(RFP), eliciting applications for the UPA program. The RFP was a bound-ary object that conveyed meaning from federal officials to state and localapplicants. The RFP addressed the syntactical, semantic and pragmaticproblems inside USDOT, and then presented applicants with the need to solvethe same problems in their proposals, which the Minnesota proposal did, asdid the final agreement between Minnesota and USDOT. Understanding theproposal and RFP as boundary objects adds to our previous finding that forg-ing sound initial agreements is an important process element in cross-sectorcollaborations (Crosby and Bryson 2005a; Bryson, Crosby and Stone 2006).

Planners of stakeholder forums during the formulation of Minnesota’s UPAapplication used what Lee Munnich calls a “grasstops” approach. They invitedstakeholders from local governments, state agencies, regional government agencies, academic programs and nonprofit advocates. They did not extendtheir invitation to the grassroots – commuters or residents surrounding candidate corridors. This approach, used successfully in planning for MnPass,kept disagreements among stakeholder groups from expanding much beyondthe forums themselves. It also was probably appropriate for the tight time-line that UPA applicants had for obtaining consensus among key stakeholdersabout the contours of the proposed project. A more problematic result isthat UPA implementers will need to invest time and effort in selling theirvision to the wider population expected to use the new integrated system.

Communications technology provided important, though less visible,cross-boundary forums. Telephone conversations obviously were important;but one interviewee also noted that, without email and the ability to includeattachments, the proposal could not have been done on time. From this per-spective, the Internet and email also may be seen as media for assemblingand revising boundary objects.

Using political leadership on behalf of the vision

As noted earlier, political leadership – the making and implementing of policy decisions in formal and informal arenas – is a necessary concomitantof visionary leadership aimed at tackling complex public problems. The most important process is building a sustainable coalition that can convincedecision-makers to approve proposed policies and that can protect them during implementation. Structure and process come together in the designand use of arenas, the settings in which policy advocates vie with each otherto obtain desired decisions from executive, legislative and administrative decision-makers. In this case, several champions helped organize the necessary

coalition by crossing political boundaries at the state level to obtain high-level support for the Urban Partnership process. Because MnDOT and theMet Council decision-makers were not fully on board, champions from the university and the Citizens League met with these people to stress theopportunity represented by the Urban Partnership program. They did thesame with key legislators. The legislators, in turn, helped MnDOT and MetCouncil change their minds by threatening to submit an Urban Partnershipproposal if MnDOT and Met Council did not. Later the university and CitizensLeague champions controversially transgressed normal decision-makingpaths by putting pressure on the governor to support the emerging contoursof the Minnesota application. Ultimately, the governor and his lieutenant-governor (who also was transportation commissioner) became sponsorsonce they were assured that congestion pricing would only be applied to exist-ing high-occupancy-vehicle (HOV) lanes and shoulders. The successfulMinnesota grant application thus was backed by a coalition of congestion-fighting advocates from the academic and nonprofit sectors, along with state,regional and local officials and transportation planners. The Minnesotacoalition remained largely intact after federal officials approved its UPA project, though some initially supportive legislators began to grumble thatthey were not kept fully informed as the Steering Committee proceeded withplanning the multiple projects that composed the agreement.

Extensive coalition-building also occurred at the federal level, where pric-ing advocates like Duvall and DeCorla-Souza convinced top administratorsto put their clout behind the approach. US Transportation Secretary Peterspersuaded White House staff that the Urban Partnership program was a goodidea. Horner ensured that needed administrative decisions were alignedwithin various USDOT areas and that the program was protected from attack in formal courts. The federal sponsors and champions recognized thatthe next president could go back to a siloed approach, so they reached out to governors in order to build political support at the state level thatwould outlast the Bush administration. They reported that some states notincluded in UPA adopted the “4 T” approach; they believe that it may be“reaching critical mass in the states”. They are trying to get the integratedmultimodal approach included in next federal surface transportation re-authorization bill.

During the first year after federal officials approved Minnesota’s UPA project, the state legislature became the most important arena, because thelegislature had to approve the $55.2 million in matching funds as well as authority to impose dynamic pricing on project roadways. Here a key sup-porter was the chair of the Minnesota Senate transportation committee, whoemphasized the safety benefits of reducing congestion and helped convinceskeptical legislators that the program was worthwhile even if it did not directlyimprove transportation in their districts.

Lessons

Analysis of integrative visionary leadership in the case of Minnesota’sUrban Partnership Agreement produces several lessons for those who aspireto practice integrative leadership.

1. Enough shared agreement about a problem permits diverse stake-holders to collaborate on remedying it, but diverse frames are still likelyto need reconciling.

2. Academic champions are important. They have access to expertise outside government agencies and can be somewhat neutral conveners and analysts who have more freedom to espouse good, evidence-basedideas.

3. A well-defined competition is an effective way to elicit good ideas thataccomplish the organizers’ goals.

4. A competitive process that promotes flexibility and local adaptation withinan overall strategic framework makes collaboration or partnership moreattractive to the local partners.

5. Advanced technology can be an important solution component but also a magnet for participation.

6. Substantial funding provides a strong incentive and may provide legitimacy.

7. The prospect of having substantial impact may help partners sign ondespite substantial risk.

8. Solution ideas should be well developed so they are ready for prime timewhen a window of opportunity opens.

9. Integrative leaders may be wise to link a vision for change to what stake-holders already want or intend to do. They should consider integratingcollaborative structures and major system innovation with regular routines and decision-making channels, so that the changes are not interpreted as renegade endeavors. They are likely to need to blend collaboration and hierarchy.

10. Cross-boundary forums are crucial. Leaders can take advantage ofexisting ones to push ideas and create new ones to develop sharedunderstanding of a problem and potential solutions as well as commit-ment to agreed-upon actions.

11. The production of boundary objects and agreement on the structure ofneeded cross-boundary groups and organizations seem to be importantoutcomes of cross-boundary forums.

12. A grasstops approach has advantages for developing solutions to complex public problems, but at some point leaders will have to focuson developing needed shared understandings and commitments at thegrassroots, too.

Conclusions

In keeping with our prior research, we identified a number of sponsors andchampions who provided essential visionary leadership in crafting a so-far-successful collaboration to remedy a complex public problem. Their partiallyshared agreement on the nature of the problem and its causes helped bridgedifferences among stakeholders from different levels of government and non-governmental organizations. They did apply a variety of competing framesto potential problem solutions. The integrative frame pushed by the federaladvocates ultimately dominated because it was backed up by substantial fund-ing and a promise of substantial impact, and because it offered ways forregional and local stakeholders to see how their pre-existing projects mightfit within it. The technology requirement made the Urban Partnership pro-gram exciting to potential partners. Still the Minnesota Urban Partnershipapplication might have foundered if pricing advocates had not succeeded inwinning over the governor with a combination of using high-occupancy andshoulder lanes to meet the tolling requirement and if federal officials hadnot been flexible enough to allow this despite its being an “impure” form ofcongestion-pricing.

This case underscores important ways that visionary leaders can cham-pion new and improved ideas for dealing with a public problem. Tying theideas to effectiveness and efficiency arguments appealed across stakeholderdivisions. The competitive RFP process and demonstration projects were especially effective in ensuring that the best ideas surfaced and accomplishedwhat the initiators hoped they would.

The analysis of this case reveals the usefulness of understanding the roleof forums in cross-boundary integration. We expect that viewing them asboundary experiences that produce boundary objects and groups and organiza-tions may be a helpful perspective for integrative leaders. Our study of theMinnesota UPA collaboration highlights the importance of particular kindsof objects central to the process. These material artifacts have functioned as“boundary objects” (Carlile 2002, 2004; Kellogg, Orlikowski and Yates 2006).Of particular importance in the UPA case was the creation and use of bound-ary objects in the form of various proposals, agreements, maps and timelinesby key stakeholders. These objects and the process of creating them werecrucial to developing shared understandings, and the coalition needed to moveforward. The role of boundary objects thus clearly merits further investigation.

Technology was a source of innovation via its enabling of congestion-pricing and telecommuting. Intelligent systems linked elements of the visiontogether. Technology was an exciter that appealed to people across govern-mental levels and across sectors. Technology was a facilitator in allowingdisparate groups to use a common medium to assemble and revise bound-ary objects. We also conclude that further study of the role of technologyin facilitating or hampering leadership across boundaries is merited.

References

Allen, K. E. and Cherrey, C. (2000) Systemic Leadership: Enriching the Meaning ofOur Work, Lanham, Md: University Press of America.

Ansell, C. and Gash, A. (2008) “Collaborative Governance in Theory and Practice”,Journal of Public Administration Research Theory, 18 (4): 543–71.

Bryson, J., Crosby, B. and Stone, M. M. (2006) “The Design and Implementationof Cross-sector Collaborations: Propositions from the Literature”, Public Admini-stration Review, 66 (s1): 44–55.

Bryson, J., Crosby, B. and Stone, M. M. (2008) Collaboration in Fighting TrafficCongestion: A Study of Minnesota’s Urban Partnership Agreement, ITS Insti-tute Publication no. CTS 08–25, Minneapolis, Minn.: Center for TransportationStudies.

Bryson, J., Crosby, B. and Bryson, J. K. (2009) “Understanding Strategic Planningand the Formulation and Implementation of Strategic Plans as a Way of Know-ing: The Contribution of Actor-Network Theory”. International Public ManagementJournal, 12 (2): 172–207.

Carlile, P. R. (2002) “A Pragmatic View of Knowledge and Boundaries: BoundaryObjects in New Product Development”, Organization Science, 13 (4): 442–55.

Carlile, P. R. (2004) “Transferring, Translating, and Transforming: An IntegrativeFramework for Managing Knowledge across Boundaries”, Organization Science,15 (5): 555–68.

Crosby, B. C. and Bryson, J. M. (2005a) Leadership for the Common Good: TacklingPublic Problems in a Shared-power World, San Francisco, Calif.: Jossey-Bass.

Crosby, B. C. and Bryson, J. M. (2005b) “A Leadership Framework for Cross-sector Collaboration”, Public Management Review, 7 (2): 177–201.

de Geus, A. P. (1988) “Planning as Learning”, Harvard Business Review, March–April: 70–4.

Feldman, M. S., Khademian, A. M., Ingram, H. and Schneider, A. S. (2006) “Waysof Knowing and Inclusive Management Practices”, Public Administration Review,66: 89–99.

Feldman, M. S. and Khademian, A. M. (2007) The Role of the Public Manager inInclusion: Creating Commitments of Participation. Governance, 20 (2): 305–24.

Friedland, R., and Alford, R. (1991) Bringing Society Back In: Symbols, Practicesand Institutional Contradictions. In W. W. Powell, & P. J. DiMaggio (Eds.), TheNew Institutionalsm in Organizational Analysis ( ). Chicago: University of ChicagoPress.

Giddens, A. (1979) Central Problems in Social Theory: Action, Structure and Con-tradiction in Social Analysis, Berkeley, Calif.: University of California Press.

Giddens, A. (1984) The Constitution of Society: Outline of a Theory of Structuration,Berkeley, Calif.: University of California Press.

Huxham, C. (2003) “Theorizing Collaboration Practice”, Public Management Review,5 (3): 401–23.

Huxham, C. and Vangen, S. (2005) Managing to Collaborate : The Theory and Practiceof Collaborative Advantage, New York: Routledge.

Kellogg, K. C., Orlikowski, W. J. and Yates, J. (2006) “Life in the Trading Zone:Structuring Coordination across Boundaries in Postbureaucratic Organizations”,Organization Science, 17 (1): 22–44.

Kingdon, J. W. (1995) Agendas, Alternatives, and Public Policies, revised edn,Boston, Mass.: Little, Brown.

Mattessich, P., Murray-Close, M. and Monsey, B. (2001) Collaboration: What MakesIt Work, St Paul, Minn.: Amherst H. Wilder Foundation.

Metropolitan Council (2007) 2007 Metro Residents Survey, St Paul, Minn.: Metro-politan Council.

Moore, M. H. (1995) Creating Public Value, Cambridge, Mass.: Harvard UniversityPress.

Orlikowski, W. (2000) “Using Technology and Constituting Structures: A PracticeLens for Studying Technology in Organizations”, Organization Science, 11 (4): 404–28.

Shein, E. H. (2004) Organizational Culture and Leadership. San Francisco: Jossey-Bass.

Stone, D. A. (2002) Policy Paradox and Political Reason, New York: W. W. Norton.Transportation Research Board (2007) Critical Issues in Transportation, Washington,

DC: Transportation Research Board.Winer, M. and Ray, K. (1994) Collaboration Handbook: Creating, Sustaining, and

Enjoying the Journey, St Paul, Minn.: Amherst H. Wilder Foundation.Winnicott, D. W. (1953) “Transitional Objects and Transitional Phenomena”, The

International Journal of Psycho-analysis, 34 (2): 89–97.

PUBLIC GOVERNANCE AND THETHIRD SECTOR: OPPORTUNITIES

FOR CO-PRODUCTION AND INNOVATION?

Victor Pestoff and Taco Brandsen

1. Introduction

In the opening chapter to this volume, Stephen Osborne presents an editorialoverview of developments from New Public Management to the emergenceof New Public Governance. He notes the increasingly interorganizational natureof public management in a clearly plural and pluralist world, where the number of actors involved in policy formulation and implementation havegrown dramatically. This chapter will focus on one type of actor: the thirdsector. Is it good to involve the third sector in public governance? Argu-ments and counter-arguments are often based on normative assumptions rather than on empirical evidence, which remains fragmented and thin onthe ground. The literature mentions several specific qualities or contributionsthat third-sector organizations could theoretically have, as compared tostate or commercial organizations. We shall here explore the nature of thosepotential contributions.

The issue is especially relevant at a time when many countries in Europehave been searching for new ways in which to include the third sector in theprovision and governance of social services. Of course, state–third-sector collaboration is not a new phenomenon and goes back a long way. Whatwe now call the “third sector” provided public services before the state everdid, and in many countries its involvement in the welfare state dates backas far as the nineteenth century. Public governance implies that multiple par-ties are involved in the delivery of healthcare, care of the elderly, education,housing, welfare, safety and other public goods. Recent decades have beenmarked by a rearrangement of the relationships between the state and third-sector organizations that supply public goods and services like education,housing, healthcare and community services. New third-sector organizationshave been drawn into public service provision; in other words, there has been

an increase in co-management between third sector and state. In addition,those third-sector organizations already in the public services have faced the challenges of quasi-marketization and/or performance measurement.Long-term relationships based on trust have been replaced by short-term,contract-based relationships, changing the nature of the government–third-sector partnership.

At a general level, the reasons for involving the third sector in public service delivery are similar throughout Europe. First is the challenge of anaging population, second is the growing democracy deficit at all levels – local,regional, national and European – and third is the semi-permanent auster-ity in public finances. In any given EU member state, the reasons will varyand may be more specific. However, taken together, they imply a major legitimacy crisis for the public sector as a provider of welfare services. It isin this context that the third sector came back into the spotlight as a providerof public services in welfare states where it traditionally did not have a majorrole; in those where it did, its role has been changing. As a result, third-sector research has become increasingly intertwined with public managementresearch, witnessed by various publications on the topic in the relevant journals and book series.

Another reason for the surge of interest in the third sector, paradoxically,is questions over its distinctiveness. The third sector comes under variousother names, such as the voluntary sector, the (private) nonprofit sector, thesocial economy, civil society, all with slightly different defining character-istics and with a large degree of overlap. When we here refer to the thirdsector, we include all those groups and organizations grouped under otherlabels, accepting that it is a “loose and baggy monster” (Kendall and Knapp1995) without trying to cage it in. This has not deterred politicians from various backgrounds from embracing the sector as one of the cures for thewelfare state’s ills. Yet the third sector is (in the context of public services)increasingly hard to get into focus. As a result of its involvement in publicservice delivery, and the contracting out and performance measurement that followed it, the traditional boundaries between market, state and thirdsector have been breaking down to the point where a class of indeterminateorganizational hybrids has emerged (Evers 2005; Brandsen et al. 2005). Itmeans that the third-sector organizations have taken on more character-istics of state organizations (e.g. in terms of formalization) and of marketorganizations (e.g. maximizing their income, but without maximizing theirprofit). Although no organization can be regarded as “pure”, many organ-izations now reach the point where the ideal types – state, market or thirdsector – no longer help us truly to understand them.1

Within this context, we shall explore what role the third sector can playin the public services. None of these roles is exclusive to the third sector,and there are good grounds to challenge whether it has a specific contribu-tion to make – indeed, whether the third-sector concept is truly useful. This

chapter will argue that, if we are to judge the distinct contribution of thethird sector to service delivery, an analysis should take account of the insti-tutional framework within which it operates. Blanket statements on the specificnature of the third sector tend to be simplistic, and this is why further comparative work is needed. In the current chapter, we shall lay out somebasic concepts along which a comparative analysis could be organized.

In the next section, we shall discuss the state of the art in research on thistopic and conclude that there are two possible benefits of third-sector involve-ment in public service delivery. We examine the first in section 3, exploringthe concept of co-production and its potential to democratize service delivery.In section 4 we go on to describe the potential contribution of the third sector, which is as a generator of innovation within the structure of serviceprovision. In both cases we shall illustrate our point with empirical evidence,though we should emphasize that we do not claim to present a comprehen-sive literature review. The chapter ends with suggestions for future research.

2. The different roles of the third sector

2.1. The state of the art

Various theoretical traditions have addressed the role of the third sector inpublic service delivery, each with its own strengths and drawbacks. We shallhere briefly describe the three main traditions: third-sector research, publicmanagement research and comparative welfare state research, including references to some key literature on the topic.

What is called “third-sector research” is a very mixed bag. Assumptionsand theories about the specific strengths of third-sector organizations in rela-tion to service delivery have been manifold. They generally concern four roles:(a) in community integration; (b) in giving the respective groups a voice; (c) in pioneering innovations in service provision that address groups, situ-ations and/or needs neglected by states and markets; (d ) their complementary role in enhancing the qualities of established public services (Zimmer andStecker 2004; Evers and Laville 2004; Nyssens et al. 2006). Past research has demonstrated convincingly at the macro-level that state–third-sector relationships may benefit both sides (Gidron et al. 1992; Salamon 1995), andthere is now a diverse body of theory on the topic (Smith and Grønbjerg2006). There is also a good working knowledge on specific issues of third-sector management, such as the management of volunteers and board composition (Cornforth 2003). However, third-sector research has been lesssuccessful in demonstrating that, within these arrangements, the distinctivestrengths of the third sector have really been addressed. Has its output differed qualitatively from that of other types of organizations?

Research in public management has by now examined quite extensivelyhow service delivery by non-governmental organizations is organized and how

PUBLIC GOVERNANCE AND THE THIRD SECTOR

this affects the nature of the relationships. For example, there is work onthe effects of national compacts, the effects of contracting out services, andthe role of these organizations in networks of service provision (Brandsen2004; Osborne 2008). There has also been much attention to specific issuesof concern to public management within the context of third-sector involve-ment, such as accountability (Kumar 2003), innovation (Osborne et al. 2008)and partnership (Carmel and Harlock 2008). This literature has been usefulin clarifying the dynamics of government–third-sector relationships, with moreattention to processes at the meso- and micro-level. Yet, as in third-sectorresearch, only some of it has been explicitly comparative (e.g. Bode 2008),between sectors and/or between welfare systems, which makes general state-ments on the functions of the third sector rather tricky.

Mainstream welfare-state research has for the past decades largely ignoredthe third sector, as it has come to be dominated by the concept of welfareregimes (Kuhnle and Selle 1992). This has been beneficial in encouraging com-parative research, but the third sector was left out of the original typologiesformulated by Titmuss that were subsequently revised by Esping-Andersen(cf. Titmuss 1974; Esping-Andersen 1990). The Esping-Andersen typologydistinguished liberal, conservative and social-democratic welfare states.Research on regimes tends to focus on national income transfer, whereas mostgovernment–third-sector partnerships concern services at the local level.Finally, the regime approach has mostly disregarded the issue of service quality, where the third sector’s distinctive characteristics are more likely tobe relevant. In short, what this tradition of research leaves us is a good under-standing of the macro-context with rather implicit assumptions about whythird-sector inputs are not so important, but little of the micro-dynamics(Pestoff et al. 2006).

Each of these traditions can contribute to an analysis of the functions of the third sector in relation to public service delivery. As noted above, several potential benefits of third-sector involvement emerge from the literature. Please note that we are not necessarily stating that these benefitsalways occur and that involving the third sector is always good. That wouldbe an altogether normative perspective. Our purpose is to specify the supposed benefits and encourage research into the conditions under whichthey materialize.

On the basis of previous third-sector research, one could roughly classifythe possible benefits of involving the sector in public service provision in termsof two main functions: democratization and innovation. This is where third-sector organizations may have an edge over their public or commercial counterparts. However, one cannot simply transplant the abstract functionsas posited by third-sector research to public governance. They need to bereinterpreted within the specific context of public service delivery, where theytake on a meaning that overlaps, but is not necessarily the same as, that withina purely voluntaristic context. For instance, the third sector supposedly

contributes to the good functioning of democracy, but what does that meanwhen it is working within a system governed by representative democraticinstitutions? Or, if certain organizations improve their services for their particular clients, and differences in quality emerge, how does that fit in systems where everyone is entitled to services of the same quality?

2.2. Co-production and collective innovation

This is why we have reinterpreted the potential contributions of the third sec-tor in two ways. Democratization will be understood in terms of co-production– direct participation by citizens in service delivery. Innovation will here beunderstood as the ability to renew the collective structure of service provi-sion, whether it be in terms of skills, activities or even the underlying paradigm.These are the two distinctive contributions to service delivery identified bythird-sector research.2 Public management research has helped to under-stand better the effect of government policy on each of these two types ofcontributions. Comparative welfare state research can help us to under-stand, within their macro-institutional context, which institutional conditionsfavor which functions, and where does the third sector have which role. Therole of citizens and of the third sector will vary among welfare regimes, with their different emphasis on individual or collective provision of socialservices and with different policies that focus on public-, private- or third-sector provision of welfare services. Let us now explore this variation, in relation to the two functions of co-production and innovation.

Co-production is one of several mechanisms that can be used to increasethe influence of citizens over the services that are delivered to them. Evers(1998) argues that there are four concepts for strengthening the position of users in service delivery. They are: (a) representative political democracy,(b) participative democracy, (c) consumerism, and (d ) involvement of the co-producer. The first concept relies on the indirect power of citizens as voters and their elected representatives. This refers to shaping the service sector according to the interests of society as a whole, not for a particulargroup. The second concept stems from a long history of self-organization,linked to the church, cooperatives, the labor movement, and other types of self-help groups and new social movements. This helps to offset the limits of parliamentary democracy and reinvigorate established voluntary organizations providing social services. There might, however, be a conflictbetween representative and participative democracy, and between a publicinterest with its universal rules and the concerns of particular groups striv-ing for a specific solution for their needs. The third concept, consumerism,claims to bring more democracy through market-like arrangements in socialservices. Given time limits, it can provide a quick fix, rather than requiringcitizen involvement in providing public services. Finally, it is the involvementof co-producers, like parent participation in schools and day care, which

promises to empower consumers and reduce the gap between the professionalsand their clientele (ibid.: 43–6).

Evers argues for a pluralistic or mixed approach of using several differ-ent pathways for democratizing the delivery of services (ibid.: 47–50). Allwelfare regimes and government policy can in their own way facilitategreater citizen participation and a greater role for the third sector in the pro-vision and governance of social services. Yet the differences between welfareregimes and differences in the sectoral context are important in how this isdone and where the role of the third sector is most distinctive – in other words,where it has a clear added benefit in relation to democratization. This is par-ticularly the case in social-democratic welfare states where the third sectoris least integrated in the established system of public service delivery as compared to its position in other regimes, and therefore paradoxically in abetter position to be responsive to specific demands than governments.

Innovation, the ability to bring something new to service delivery, has been one of the chief justifications for involving the third sector. But what isinnovation exactly? We shall here define it as a significant change in the pro-cess of the “production” of services. Whereas co-production refers to whois involved in the process, innovation concerns the qualities (in a neutral sense)of the service itself. It is of course possible to distinguish different kinds of innovation, depending, for instance, on the aspects of the service that arechanged or the structure of the process of change.3 We shall here keep matters simple, for reasons of space. We should also note that innovationis not necessarily always good: sometimes it is better to maintain the statusquo and not allocate resources to changes that are not needed.

An important point is that innovation itself is not sufficient in the context of public services. There must also be a process that makes such innovations accessible to a broader range of users. After all, the major dis-advantage of the third sector (at least from the perspective of public servicedelivery) is its particularistic nature, restricting improved quality to a selectgroup of users. Indeed, this was historically one of the chief arguments for supporting a greater role of the state in financing and delivering welfarestate services. In other words, whether innovation is a useful function of third-sector organizations depends not only on the performance of single organizations, but also on the ability of the overall field of organizations to transfer and adopt it. We shall therefore conceptualize innovation as acollective process.

We shall now proceed to discuss these two themes in more detail, describ-ing the basic concepts and sketching some of the empirical evidence relatingto them. Please note again that we are not making prior assumptions concerning the value of the third sector’s contribution, but rather whittlingdown the variety of its potential contributions toward manageable and measurable concepts, for the systematic empirical research on the topic thatis so badly needed.4

3. Co-production: citizen participation in service delivery

3.1. The co-production concept

Hirst (2002) argued that big organizations on either side of the public/private divide in advanced post-industrial societies leave little room fordemocracy or citizen influence. This is due to the lack of local control anddemocratic processes for internal decision-making in most big organizations.Evers (2006) maintained that user involvement in welfare services is a general concern throughout Europe and that there are at least five differentapproaches to involvement. They are partially overlapping and partiallyconflicting. They range from welfarism and professionalism, through con-sumerism and managerialism to what he calls participationalism. They arebased on different values and promote different degrees of user involvement.These approaches will vary among sectors and over time. Their mix will probably differ among countries. Welfarism and professionalism are closelyassociated with each other, and neither leaves much room for user involve-ment. Rather, clients are viewed as people with little competence of their own. Consumerism and managerialism call for giving users greater choicethrough more exit options, and argue that the public sector needs to learnfrom the private sector (ibid.). However, they leave little room for voice or participation.

Participationalism encourages on-site participation by users of welfare services, based on the belief that citizens should engage personally in shaping the welfare services they demand. It emphasizes multi-stakeholderorganizations and requires that users become co-producers. Welfarism and professionalism are usually promoted by social democratic governments,while consumerism and managerialism are normally championed by rightistgovernments. However, participationalism – or, more simply, co-production– lacks clear political proponents in most EU countries. In a service demo-cracy of either the social democratic or rightist variety, citizens are the con-sumers of public-financed social services provided by municipal authorities,regional governments and private companies. They vote every fourth yearand in the mean time they choose between various public or private serviceproviders. This contrasts with views of participative democracy in which citizens are engaged in the provision of some of their own social services.

This is the background against which the concept of co-production has developed. The notion of citizen involvement in the provision of public services generated much interest among public administration scholars inAmerica in the 1970s and the 1980s (see Parks et al. 1981 and 1999 for agood overview). It experienced somewhat of a revival with the publicationat the turn of the century of a paper on incentives structures for co-productionin Australian public services (Alford 2002) and in recent work on the Swedishwelfare state (notably Pestoff 2006; Vamstad 2007). It was originally developed

by the Workshop in Political Theory and Policy Analysis at Indiana Uni-versity. During the 1970s they struggled with the dominant theories of urbangovernance underlying policy recommendations of massive centralization.Scholars and public officials argued that citizens as clients would receive more effective and efficient services if they were delivered by professional staff employed by a large bureaucratic agency. But this group of researchersfound no empirical support for such claims promoting centralization (Ostrom1999: 358).

They did, however, stumble on several myths of public production. Onewas the notion of a single producer being responsible for urban services withineach jurisdiction. In fact, they normally found several agencies, as well asprivate firms, producing services. More important, they also realized that the production of a service, in contrast to goods, was difficult without theactive participation of those receiving the service. They developed the termco-production to describe the potential relationship that could exist betweenthe “regular” producer (street-level police officers, schoolteachers, or healthworkers) and “clients” who want to be transformed by the service into safer,better-educated or healthier persons. In complex societies there is a divisionof labor, and most persons are engaged in full-time production of goods andservices as regular producers. However, individual consumers or groups of consumers may also contribute to the production of goods and services,as consumer-producers. This mixing may occur directly or indirectly. Co-production is, therefore, noted by the mix of activities that both public service agents and citizens contribute to the provision of public services. Theformer are involved as professionals or “regular producers”, while “citizenproduction” is based on voluntary efforts of individuals or groups toenhance the quality and/or quantity of services they receive (Parks et al. 1981/1999). Co-production is one way in which a synergy could occur betweenwhat a government does and what citizens do (Ostrom 1999).

3.2. Empirical evidence

There is a lot of work on citizen participation, but relatively little in the context of service delivery. The TSFEPS Project5 examined the relationshipbetween parent participation in the provision and governance of childcarein eight EU countries (Pestoff 2006/2008). It found different levels of par-ent participation in different countries and in different forms of provision,i.e. public, private for-profit and third-sector childcare. The highest levels of parent participation were found in third-sector providers, like parent asso-ciations in France, parent initiatives in Germany and parent cooperatives in Sweden. Different kinds of parent participation (i.e. economic, politicaland social) were readily evident in third-sector providers of childcare services, while both economic and political participation were highly restrictedin municipal and private for-profit services (ibid.). Later, Vamstad (2007)

confirmed the existence of these three dimensions of co-production in Swedishchildcare and underlined clear differences between types of providers con-cerning the saliency of these dimensions in providing welfare services (Pestoff2008). This provides parents with unique possibilities for active participa-tion in the management and running of their child(ren)’s childcare facilityand for unique opportunities to become active co-producers of high-qualitychildcare services for their own and others’ children. It is also clear that other forms of childcare allow for some limited avenues of co-production inpublicly financed childcare, but that the parents’ possibilities for influencingthe management of such services remain rather limited.

In other words, existing empirical work appears to indicate that parti-cipation by individual citizens is best-facilitated by third-sector providers. What is less clear is why exactly this is the case and under what conditionsco-production is most likely to be effective. So far, research on the topic hasbeen conducted primarily (though not exclusively) in the context of social-democratic welfare regimes where service delivery is dominated by state provision. In such regimes, the third sector is least integrated within the structure of public service provision, which is arguably why co-productionappears to function best within third-sector organizations. Sweden seems tobe fertile ground for co-production (interestingly, this is the same countrywhere the IKEA model of production originated, although there is no obviousconnection). The objective for future comparative research is to determinehow well co-production functions in other kinds of welfare-state regimes.

4. Innovations in public service delivery

4.1. The nature of innovation

Earlier, we defined innovation as a significant change in the process of production. A lot of third-sector literature appears to assume that the organizations in this domain are innovative by nature. After a review of theliterature, Osborne (1998) concluded that studies on this topic relied on normative argument rather than on empirical data, were insufficiently awareof mainstream innovation-studies literature and failed to understand the contingencies of innovation, especially the impact of the public policy envir-onment upon innovativeness. He argued that innovation emerges from theinteraction of third-sector organizations with their institutional and pol-icy environments. As we shall show, there is some evidence to support thelatter claim.

A favorable institutional environment may be necessary not only toengender innovation within particular organizations but also to make surethat it spreads across to all other organizations. In this context, is worth turn-ing to DiMaggio and Powell’s (1991) classic study on isomorphism. Theysuggested that, once organizations are structured into a field, forces come

into play which encourage similarity. In each organizational field there area few true innovators who have the desire and the capacity to improve theirperformance by coming up with innovations. Other organizations will copythe ideas of these innovators, rather than innovate themselves. Particular innov-ations will thus gradually spread throughout the field. The aggregate effectof change will be diminished diversity, if not outright homogeneity, in thecommunity that shares the innovations. One of the mechanisms through whichthis drive toward homogeneity occurs is imitation. Organizations volun-tarily copy strategic elements from other organizations in their field (“mimeticisomorphism”). Another way is for people with similar educational back-grounds to come to dominate within fields of organizations (“normative isomorphism”). As they become more widespread, the innovations becomemore legitimate simply because they are widespread. Finally, the state or otherpowerful bodies in the environment may force organizations to adopt par-ticular changes (“coercive isomorphism”). Whether and how such mechanismsaffect the third sector is a significant empirical question, because, as far asinnovation is concerned, it determines their significance to the public sector.

4.2. Empirical evidence

There is little systematic empirical work on the innovative capacity of third-sector organizations. The study by Osborne which we quoted earlier examined organizations in social welfare in the UK and found that the “policy context created by central and local government encouraged innov-ative activity by VCOs rather than it being an inherent consequence of theirorganizational structure or culture” (Osborne et al. 2008). Public policy wasan important determinant of the extent and nature of innovation. Futureresearch should further investigate the conditions under which governmentpolicy can be conducive to innovation. Research from the Netherlandsshows evidence of a link between hybridization and innovation. As service-providers become more hybrid, and given that they have sufficient autonomy,they appear to start taking initiatives to reconcile the tension between theirmarket, state and community characteristics. Such initiatives may be pre-sented as “a focus on communities”, “demand-driven supply”, “putting thecustomer central”, borrowing from both market or third-sector discourse,or both. In terms of organizational theory, it comes down to a strategy ofdiversification, possibly (though not necessarily) accompanied by a cognitivechange in how the problems they need to solve are framed. This implies that organizations start to expand into new areas of activity where they believedtheir clients could be served better. Such diversification may even lead to a revision of the organizational mission. In an earlier article (Brandsen and Van Hout 2006) we examined the example of social housing, where social landlords gradually moved from a classical role of merely renting out their property on the basis of individual contracts to being leading

members in programs of urban regeneration and community investment. Theywent beyond their original (bureaucratically defined) function and adopteda more holistic perspective, in which they would “revitalise communities” or something similar, treating their housing stock as a means rather than an end. This diversification may constitute an innovation at the level of thestructure of service provision, if it is adopted throughout the field.

The second dimension of innovation concerns whether such strategies can spread beyond single organizations. In the context of public services, itis obvious that governments play an important role in disseminating innov-ation, by imposing (“mainstreaming”) desired changes in service provisionthrough regulation. The role of coercive isomorphism hardly needs mentioningin this context. There are theoretical reasons to suppose that the other typesof isomorphism are less strong in the third sector than among other types oforganizations, because it generally has a lower level of professionalizationand is less tightly organized. However, they may be at play where third-sector organizations are tightly integrated in local and/or functional networksof service delivery. If innovation is interpreted as a collective effort, then itis the outcome of networks rather than of organizations, in the same sensethat it would be for complex technological innovation (Powell et al. 1996).For instance, a case study in social housing in the Netherlands demonstratedthat, although most organizations lacked a basic capacity to develop new ideas,the organizational field as a whole could progress because innovations byfrontrunners were massively copied (Brandsen 2004). Within such a type ofanalysis, the process of innovation is conceived of as an interaction betweenorganizations and the networks to which they belong ( just as co-productionemerges from the interaction of the organization and individual citizens).

If integration into public service networks is conducive to isomorphism,then the third sector has been most important to innovation in those insti-tutional regimes where it is best-embedded and where it traditionally playsa pivotal role. This applies primarily to Bismarckian or conservative welfarestates and to a lesser extent to Anglo-Saxon ones (even though the role ofthe third sector in the latter has often been understated).

5. Conclusion

In public governance, the third sector has come to play a more importantrole in public service delivery. Yet, despite many popular assertions, thereis as yet only limited empirical evidence on the actual contribution of thethird sector. This is not to say that it has none, simply that the availablematerial refers to case studies (e.g. Titmuss’ famous work from 1970 on blooddonations) and that few studies have systematically analyzed the perform-ance of the third sector under different conditions and in comparison withother types of providers. Such work as there is shows mixed results. It isimperative to continue and expand these studies over the coming years.

The conditions for setting up such a comparative program are better than they have been for a long time. There are now various third-sector research centers throughout Europe, which are connected through variouscross-national research networks. In public management research, the thirdsector has become an established subtheme and one that benefits from theinterest in new public governance. This will hopefully result in systematiccomparative research into the benefits and drawbacks of the third sector’sinvolvement in public service delivery, and perhaps in a debate that is basedmore on evidence and less on normative assumptions.

1 Indeed, some authors have defined the third sector as essentially hybrid, anindefinable area that as yet lacks a distinct identity (Brandsen et al. 2005). It couldeven be argued that the third sector is better-understood in terms of certainaspects of public services, rather than as a distinct cluster of organizations (Evers2005; Brandsen et al. 2009).

2 It is clear that these two benefits can be empirically related, but they differ conceptually in that the former relates to the community as the primary object ofanalysis, whereas the latter is foremost about the system that produces services.

3 Osborne (2008) distinguishes between total innovation (involving working with anew client group and providing new services), expansionary innovation (involvingworking with a new client group, but using the existing services/methods of work of the organization), evolutionary innovation (involving working with thesame client group, but providing new services), and incremental development(involving working with the same client group and providing the same services,but incrementally improving them).

4 It might be argued that our presentation could be strengthened by separating the analysis of the third sector in instrumental terms as a service deliverer fromparticipatory approaches, where the third sector can perhaps enhance democraticparticipation. Each approach could then be presented independently. However, we feel that these two approaches are often implicitly or explicitly linked and sometimes inseparable. Therefore, we introduce each concept separately and presentsome initial empirical evidence for each within the same presentation. We deem it feasible for readers to distinguish between the two; and choose, therefore, notto present these two streams of analysis separately in independent chapters, for reasons of space and clarity.

5 The TSFEPS Project, Changing Family Structures and Social Policy: ChildcareServices as Sources of Social Cohesion, took place in eight European countriesbetween 2002 and 2004. See www.emes.net for details and reports. The eight countries were: Belgium, Bulgaria, England, France, Germany, Italy, Spain andSweden.

References

Alford, J. (2002) “Why Do Public Sector Clients Co-produce? Towards a Contin-gency Theory”, Administration and Society, 34 (1): 32–56.

Bode, I. (2008) The Culture of Welfare Markets: The International Recasting of Pensionand Care, London/New York: Routledge.

Brandsen, T. (2004) Quasi-market Governance: An Anatomy of Innovation, Utrecht:Lemma.

Brandsen, T., Dekker, P. and Evers, A. (2009) Civicness in the Governance and Deliveryof Social Services, Baden-Baden: Nomos.

Brandsen, T., van de Donk, W. and Putters, K. (2005) “Griffins or Chameleons?Hybridity as a Permanent and Inevitable Characteristic of the Third Sector”,International Journal of Public Administration, 28 (9–10): 749–65.

Brandsen, T. and van Hout, E. (2006) “Co-management in Public Service Networks:The Organisational Effects”, Public Management Review, 8 (4): 537–49.

Carmel, E. and Harlock, J. (2008) “Instituting the ‘Third Sector’ as a Govern-able Terrain: Partnership, Procurement and Performance in the UK”, Policy andPolitics, 36 (2): 155–71.

Cornforth, C. (2003) The Governance of Nonprofit and Voluntary Organisations,London: Routledge.

DiMaggio, P. J. and Powell, W. W. (1991b) “The Iron Cage Revisited: InstitutionalIsomorphism and Collective Rationality in Organizational Fields” in W. W.Powell and P. J. DiMaggio, eds., The New Institutionalism in OrganizationalAnalysis, University of Chicago Press, pp. 63–82.

Esping-Andersen, Gösta (ed.) (1996) Welfare States in Transition: National Adapta-tions in Global Economics, London/Thousand Oaks, Calif./New Delhi: Sage.

Evers, Adalbert (1998) “Consumers, Citizens and Coproducers – a PluralisticPerspective on Democracy in Social Services”, in Gaby Flösser and Hans-Uwe Otto(eds) Towards More Democracy in Social Services: Models and Culture of Welfare,Berlin/New York: Walter de Gruyter.

Evers, A. (2005) “Mixed Welfare Systems and Hybrid Organizations: Changes in the Governance and Provision of Social Services”, International Journal of PublicAdministration, 28 (9–10): 737–48.

Evers, A. (2006) “Complementary and Conflicting: The Different Meaning of ‘UserInvolvement’ in Social Services”, in Aila-Leena Matthies (ed.) Nordic Civic SocietyOrganizations and the Future of Welfare Services: A Model for Europe?, Copenhagen:Nordic Council of Ministers, TemaNord.

Evers, A. and Laville, J.-L. (eds) (2004) The Third Sector in Europe, Cheltenham:Edward Elgar.

Gidron, B., Kramer, R. M. and Salamon, L. M. (1992) “Government and the ThirdSector in Comparative Perspective: Allies or Adversaries?”, in B. Gidron, R. M.Kramer and L. M. Salamon (eds) Government and the Third Sector: EmergingRelationships in Welfare States, San Francisco, Calif.: Jossey-Bass.

Hirst, Paul (2002) “Democracy and Governance”, in Jon Pierre (ed.) DebatingGovernance, Authority, Steering and Democracy, Oxford: Oxford University Press.

Kendall, J. and Knapp, M. (1995) “A Loose and Baggy Monster: Boundaries,Definitions and Typologies”, in J. D. Smith, C. Rochester and R. Hedley (eds) An Introduction to the Voluntary Sector, London: Routledge, pp. 66–95.

Kuhnle, S. and Selle, P. (1992) “Government and Voluntary Organizations: ARelational Perspective”, in S. Kuhnle and P. Selle (eds) Government and VoluntaryOrganizations, Aldershot: Avebury.

Kumar, S. (2003) Accountability: A Qualitative Study of Relationships between thePublic Sector, the Voluntary Sector and Users of Health and Welfare Services inthe Context of Purchase of Service Contracting, Birmingham: University of Aston.

Nyssens, M., Adam, S. and Johnson, T. (2006) Social Enterprise: At the Crossroadsof Market, Public Policies and Civil Society, London: Routledge.

Osborne, S. P. (1998) Voluntary Organizations and Innovation in Public Services, London:Routledge.

Osborne, S. (2008) The Third Sector in Europe: Prospects and Challenges, London/New York: Routledge.

Osborne, S. P., Chew, C. and McLaughlin, K. (2008) “The Once and FuturePioneers? The Innovative Capacity of Voluntary Organisations and the Provisionof Public Services: A Longitudinal Approach”, Public Management Review, 10 (1):51–70.

Ostrom, E. (1999) “Crossing the Great Divide: Coproduction, Synergy, and Develop-ment”, in Michael D. McGinnis (ed.) Polycentric Governance and Development:Readings from the Workshop in Political Theory and Policy Analysis, Ann Arbor,Mich.: University of Michigan Press.

Parks, Roger B. et al. (1981/1999) “Consumers as Co-producers of Public Ser-vices: Some Economic and Institutional Considerations”, Policy Studies Journal,9: 1001–11; and reprinted in Michael D. McGinnis (ed.) Local Public Economies:Readings from the Workshop in Political Theory and Policy Analysis, Ann Arbor,Mich.: University of Michigan Press.

Pestoff, Victor (2006/2008) “Citizens as Co-producers of Welfare Services: Childcarein Eight European Countries”, Public Management Review, 8 (4): 503–20; andreprinted in Victor Pestoff and Taco Brandsen (eds) Co-production: The Third Sectorand the Delivery of Public Services, London/New York: Routledge.

Pestoff, V., Osborne, S. and Brandsen, T. (2006) “Patterns of Co-production in PublicServices: Some Concluding Thoughts”, Public Management Review, 8 (4): 591–5.

Powell, W. W., Koput, K. W. and Smith-Doerr, L. (1996) “Interorganizational Col-laboration and the Locus of Innovation: Networks of Learning in Biotechnology”,Administrative Science Quarterly, 41: 116–46.

Salamon, L. M. (1995) Partners in Public Service: Government–Nonprofit Relationsin the Modern Welfare State, Baltimore, Md/London: Johns Hopkins UniversityPress.

Smith, S. R. and Grønbjerg, K. A. (2006) “Scope and Theory of Government–Nonprofit Relations”, in W. W. Powell and R. Steinberg (eds) The Nonprofit Sector:A Research Handbook, 2nd edn, New Haven, Conn.: Yale University Press.

Titmuss, R. M. (1974) Social Policy: An Introduction, London: George Allen & Unwin.Vamstad, Johan (2007) “Governing Welfare: The Third Sector and the Challenges to

the Swedish Welfare State”; Östersund: Ph.D. Thesis, No. 37.Zimmer, A. and Stecker, C. (2004) Strategy Mix for Nonprofit Organisations: Vehicles

for Social and Labour Market Integration, New York: Kluwer/Plenum.

Part III

GOVERNANCE OF CONTRACTUAL

RELATIONSHIPS

GOVERNANCE, CONTRACTMANAGEMENT AND PUBLIC

MANAGEMENT

Donald F. Kettl

In the late 1970s and early 1980s, citizens throughout the world pressed theirelected officials to shrink the size of government. It did not matter what theactual size of government was. From the large Scandinavian welfare statesto the relatively smaller governments in the United States, the government-limitation movement ignited and spread (Pollitt and Bouckaert 2000).Citizens, of course, had little interest in reducing the services they received.They wanted substantial government programs but they wanted to spend less, because they were convinced that government had become too bloatedand inefficient.

That created a dilemma for government officials: how to avoid enragingcitizens by cutting services they liked, and how to devise new strategies toshrink government’s size and cost. The central strategy was privatization:selling enterprises to the private sector, where they could find buyers, andexpanding the contracting out of goods and services, where they could findwilling partners (Savas 2000). Some governments sold airlines, railroads,telecommunications companies, utilities and the post. Other governmentsexpanded their use of contracting out and other partnerships with the private and nonprofit sectors (Schick 1996; English and Guthrie 2003;Albalate, Bel and Fageda 2007). Getting a good fix on this vast collectionof activity is notoriously difficult. Most governments do not keep good recordson the variety or volume of activity that is outsourced, and what little datathere are are often not comparative over time. For example, in the UnitedStates, the Federal Procurement Data System tracked contract data, but the database was often unreliable.

Of all the strategies, contracting out poses the most fundamental, ongoinggovernance and public management puzzles. Once a government sells an enterprise, its governance responsibilities do not end. Taxpayers might haveseller’s remorse about the good old days when taxpayers supported services.More important, governments typically retain a regulatory role in ensuring

the quality and safety of the services. In the United Kingdom, for example,balancing government regulation of the railroads with relying on the profitmotive has been an ongoing struggle. Separating rail infrastructure and main-tenance has proved especially problematic (Preston 1996). Although criticsargued that the UK reforms had undermined safety, there is no evidence that overall safety has deteriorated (Evans 2007). Heavy pressure to meetschedules tended to undercut long-term maintenance and, critics argued, managing government contracts poses the most complicated ongoing puzzlebecause it brings in fundamental boundary challenges – what is public andwhat is private – and a different combination of management puzzles – howthe public sector can gain leverage over the performance of private organ-izations (Osborne 2002).

It is tempting to point to the government-limitation movement as the water-shed of these challenges. In fact, contracting out stretches back millennia.Caesar’s chronicles of his campaigns are full of tales of managing the procurement of supplies for his troops. He could, of course, have relied onan extended supply-line from Rome to the furthest reaches of the empire,but that would have been extremely difficult to create and maintain. Backedby the strongest army in the world, ready to take what it could not buy,Caesar found it easier simply to procure what he needed from the locals.Machiavelli wrote about the outsourcing of an entire army. Sometimes thiscame at the point of a lance, but it was part of a millennia-old governmentreliance on procurement from private suppliers. It has been called by a wide variety of names, from “contracting out” to “competitive tendering” to“outsourcing”, but interest in the movement is strong and growing. Not onlyare governments relying on private suppliers of goods and services, from paperand pencils to trucks and buses; they are quite literally sharing power withtheir private-sector partners (Kettl 1993). While the modern contracting move-ment has grown rapidly in the United States and Australia, in particular, ithas spread to many countries around the world, especially in Australia andBritain (Boston 2000; Pollitt and Bouckaert 2000).

Governments have sometimes relied on such procurement because it wasmore convenient, sometimes because it was cheaper, sometimes because itmade pragmatic sense. Powerful armies can simply take what they need; butthat is an expensive way to provision an empire, for it requires a large armyto extract the goods and services and to maintain order. In the long run, it is easier for governments to buy what they need than to take it, and overtime a body of procurement law has grown up to shape and govern the pro-curement system (Gellhorn et al. 2003).

In the United States, for example, contracting out goes back to thenation’s first days. During the American revolution, the army contracted out for arms and provisions, and General George Washington constantlycomplained that his suppliers were defrauding the government and steal-ing the higher-quality goods from his troops. In fact, throughout American

GOVERNANCE OF CONTRACTUAL RELATIONSHIPS

history, the government has relied heavily on contractors for fighting wars. In the Revolutionary War, launched against the British in 1776, therewas a contractor employee for every six soldiers (see Table 14.1). That ratio held steady for the next century of conflicts. By the war in Iraq in early 2008, there was a contractor employee for every soldier, a ratio thatproved important not only for logistical support but also for reducing political opposition. If the American army had relied on soldiers for opera-tions contracted out, six times the number of soldiers would have been required.This would have been a force unsupportable except through a large-scale draft, and it would unquestionably have stirred insurmountable political opposition.

GOVERNANCE, CONTRACT MANAGEMENT

Table 14.1 Presence of contractor personnel during US military operations

Estimated personnel (thousands) Estimated ratio

of contractor toConflict Contractor a Military military personnel a

Resolutionary War 2 9 1 to 6War of 1812 n.a. 38 n.a.Mexican-American War 6 33 1 to 6Civil War 200 1,000 1 to 5Spanish-American War n.a. 35 n.a.World War I 85 2,000 1 to 24World War II 734 5,400 1 to 7Korea 156 393 1 to 2.5Vietnam 70 359 1 to 5Gulf War 9b 500 1 to 55b

Balkans 20 20 1 to 1Iraq Theater as of Early 2008c 190 200 1 to 1

Source: Congressional Budget Office based on data from William W. Epley, “Civilian Supportof Field Armies,” Army Logistician, vol. 22 (November/December 1990), pp. 30–35; Steven J.Zomparelli, “Contractors on the Battlefield: What Have We Signed Up For?” Air Force Journalof Logistics, vol. 23, no. 3 (Fall 1999), pp. 10–19; Department of Defense, Report on DoD Programfor Planning, Managing, and Accounting for Contractor Services and Contractor PersonnelDuring Contingency Operations (October 2007), p. 12.

Note: n.a. = not available.a. For some conflicts, the estimated number of contractor personnel includes civilians employed

by the US government. However, because most civilians present during military operationsare contractor personnel, the inclusion of government civilians should not significantly affectthe calculated ratio of contractor personnel to military personnel.

b. The government of Saudi Arabia provided significant amounts of products and services dur-ing Operations Desert Shield and Desert Storm. Personnel associated with those provisionsare not included in the data or the ratio.

c. For this study, the Congressional Budget Office considers the following countries to be partof the Iraq theater, Bahrain, Jordan, Kuwait, Oman, Qatar, Saudi Arabia, Turkey, and theUnited Arab Emirates.

Source: http://cbo.gov/ftpdocs/96xx/doc9688/08-12-IraqContractors.pdf, p. 13.

The contracting-out process thus has deep roots but a modern focus.Governments rely on contracts to obtain goods and services they cannot –or do not wish to – produce on their own. They contract out to reduce theircosts and to create at least the appearance of smaller governments, becauseonly rarely do they transparently track the number of contractors or theamount of money spent through them. Contractors allow government toexpand without increasing their size – at least as measured by the numberof government employees (Light 1999). Although obtaining good numberson the contracting-out process is difficult in individual countries, andimpossible on a global scale, all signs point to a dramatic increase in the useof this indirect tool of government since the tax-limitation movement beganin the late 1970s (Pollitt and Bouckaert 2000). The contracting-out move-ment has rippled throughout the world, in part because governments havealways contracted out and in part because an expansion of the movementprovided a tactical solution to the cross-pressures governments have faced(McLaughlin, Osborne and Ferlie 2002). From more traditional governmentcontracts to purchase goods and services to the use of public–private partner-ships to fund transportation projects, the interpenetration of the public andprivate sectors is large and growing.

Public management puzzles

As contracting out has increased, however, government’s capacity to man-age the contracts has not kept pace. While many contracts have worked verywell, serious problems have emerged with many contracts. The most notablecatalog of problems comes from the “high-risk list” assembled by the USGovernment Accountability Office. The GAO identifies programs that, itbelieves, are especially susceptible to waste and mismanagement. In 2008, itshigh-risk list numbered twenty-eight programs, including the managementof contracts for defense and space programs, and the modernization of informa-tion systems for tax collection and air traffic control. In fact, virtually all of the items on the GAO’s list involve substantial contracting out – exceptfor the management areas like human capital and information technologyrequired to run the contracts (US GAO 2008a).

Moreover, many countries are reforming their regulatory strategies for over-seeing private-sector contracts. For example, as contaminants have crept intothe production chain for pharmaceuticals, the European Union and manyother governmental bodies have sought to redefine how best to regulate the quality and safety of drugs. Thus, governments are not only relying more on contracting for producing their own goods and services; they arealso finding themselves drawn inexorably into regulating the contracting processes of private companies whose products affect the public interest (Brownand Jacobs 2008).

These challenges frame a critical issue: contracts raise serious governanceproblems, and a failure to solve them brings major consequences. It is notso much that contracts are, by their very nature, harder to manage than othergovernment tools. Rather, contracts must be managed differently from othertools, like direct tools, regulations and tax expenditures (Salamon 2002). Whatworks well in managing direct government services, especially through the traditional approach of hierarchical authority, frequently does not work well at all for contracts. In direct provision of government services, superiorsrely on authority to oversee their subordinates, who work underneath themin the agency’s hierarchy. In contracts, by comparison, government officialsoversee performance through the negotiated provisions of the contract,enforced by civil law. Governments have often encountered serious prob-lems in managing their contracts because they have relied on the tools formanaging direct contracts in seeking to steer the performance of contracts(Kettl 2002). When there is a mismatch of management tools and governancetactics, problems result – much as would be the case for using a hammer todrive a screw or a saw to loosen a bolt.

Governments frequently slide into this mismatch because the ideology promoting contracting out tends to presume that contracts will be self-executing. The argument is simple. Private markets are believed superior togovernments, because of the discipline that the marketplace provides. Theforces of supply and demand provide self-correcting forces for problems ofprice and quality. So turn government’s work over to the private sector andthe markets will ensure effective and efficient services. Of course, no privatepurchaser would blindly trust the market with piles of money when buyingimportant goods and services. That is the central meaning of caveat emptor.Public antipathy toward government sometimes overpowers the lessons thatsmart private buyers know all too well. Contracts, of course, are not self-executing, and governments need to behave as smart private-sector buyersdo: to know what they want to buy, to deal in markets that will providewhat they need, and to ensure that what they get is what they wanted (Kettl1993; Padovani and Young 2006, 2008).

Moreover, private market discipline depends on competitive markets, withlarge numbers of buyers dealing with large numbers of sellers about goodswhose nature and performance are relatively straightforward. In many govern-ment services, however, these conditions of market discipline are lacking; that is, there are what economists call “market failures”. Such market failuresconstitute an age-old case for government intervention. The paradox of thecontracting-out movement, therefore, is that the expansion of contractingout, as part of a movement to shrink government, has built a case for strongergovernment action, but in a different mode for which governments sometimesare ill-prepared. As government’s contracting out grows into new areas, theseproblems of market failure multiply and bring new governance challenges.

The nature of goods and services

Governments contract for a vast array of goods and services. Some are relatively homogenous goods and services, from custodial and food servicesto automobiles and trucks. In fact, in some countries, governments have largely eliminated the jobs of government employees who used to serve asjanitors and cafeteria workers – even security staff – and have contractedout those positions. But governments also contract out for a wide variety ofnon-standard items. Contractors often pilot unmanned drones thousands ofmiles away from the battlefield while sitting next to military officials. Govern-ments rely on contractors to process the claims for government-funded healthbenefits and for programs to provide job skills to unemployed workers.

The further the contracting-out process moves from relatively standard goodsand services, the further the process moves from full market competition andthe harder it is to rely on market discipline. Private companies, of course,know this. They do not simply pay for everything their suppliers want tosell them and they do not blindly accept the low bid. In fact, for even mildlyunusual goods and services – and especially for important items on whichtheir businesses depend – private companies often seek a predictable supplyof high-quality goods as much as the lowest cost (Donahue 1989). An auto-mobile assembly line will quickly break down if the lowest bidder cannotreliably deliver seats or if windows break as workers try to install them. Thegrowth of contracting out has taken government deeper into more of thesenon-standard areas, including weapons production, environmental services,transportation systems, social services, and a host of other areas. The verynature of these goods and services has taken contractors deeper into the full range of government’s goods and services, and farther from the basicassumptions of market discipline.

The lesson is that even private companies do not negotiate contracts solely on the basis of price. These considerations of quality and reliabilityoften take government away from the underlying assumptions that drove theexpansion of government contracting, and more into issues that require detailedcontract provisions.

Monopoly

Analysts have long known that the virtues of markets quickly erode if buyershave few sellers from which to choose. For many government programs, thatis precisely the case. When the American government sought to buy a new-generation fleet of aerial refueling tankers, it had only two options: Boeingand EADS, the manufacturer of Airbus. Contractors spend 90 percent of the budget for America’s space shuttle, but NASA works through UnitedSpace Alliance, a consortium of the two biggest aerospace contractors,Boeing and Lockheed Martin, because neither company was big enough

to manage the project on its own (NASA 2003: 104). In a host of cases, from fighter planes to aerial tankers, and from submarines to social-service programs, government often has limited choice for companies providing specialized goods and services.

In fact, because many of these programs would not exist were it not forpublic purchases, government must often first create the partnerships andthen seek to manage them. Contracting out for relatively standard goods and services is one thing, but the expansion of government contracting hasnecessarily brought government into policy areas where there are relativelyfew contracts. This raises the fundamental warning of economists: if government is presumed inefficient because of its monopoly status, privatemonopolies are unlikely to be any better. After all, it is not the publicnessor privateness of service provision that, according to the argument for privatization, promotes greater efficiency. It is competition and choice.When there is little of either, government is unlikely to reap big savings andis very likely to court governance problems in countering the market failures.This does not mean that contracting out does not make sense in monopolyconditions. Indeed, government might wish to rely on contractors for expertise in areas where it lacks specialized skills or does not wish to investin building them. But, when it contracts with monopolistic suppliers, government must be on the alert for the dangers of market failure and mustadjust its own strategies to counter it (Padovani and Young 2006).

Monopsony

Economic analysts have long recognized the failures that follow from markets where there are few suppliers. But there are also risks that occur in markets in which there are few buyers. For many goods and services forwhich the government contracts out, it is the only buyer. Government is theonly purchaser of long-range bombers and attack fighters, of nuclear sub-marines and aircraft carriers, of nuclear weapons and anti-aircraft rockets(although government officials fear the creation of black markets among terrorists for some of these items). Worries about the spread of strategicallyimportant technology and about relying on suppliers in other countries forstrategic goods often further reduce the number of potential suppliers. Infunctions like the provision of social services, government is not the onlybuyer. International nongovernmental organizations like the Red Cross alsoplay a substantial role in such functions; but, as government’s contractingout has increased, in many countries it has come to dominate the marketsfor the provision of such services.

The argument for outsourcing begins with the presumed superiority of market competition. However, when government is the only buyer, privatesuppliers become dependent on government for their sales, which in turn dis-rupts market competition. Governments have no way of determining from

the market what a fair price or good performance means, because withouttheir role as buyer there would be no price, no performance and no market.For many goods and services, the market failures multiply: government isoften the only buyer and frequently must deal with a very small number ofpotential sellers for goods and services where the government marketplaceoften is the only one.

This is not to say that such imperfect markets, with limited numbers of buyers and sellers, are inherently problematic. Neither is it necessarily the case that the absence of market competition necessarily means that thegovernment should not rely on private contractors. Rather, it means thatthe government’s reliance on private markets for such goods and services ismore likely to be based on pragmatism than on low price. In World War II,for example, the US government used private contractors throughout the country to build the first nuclear weapons, because it did not wish to buildand manage the factories itself and because it wished to scatter productionto protect the facilities from attack. Instead it relied on GOCOs – government-owned, contractor-operated plants – to provide the expertise and flexibilityneeded to produce such complex systems (Kettl 2002). That arrangement grewinto an effective partnership that lasted sixty years. The relationship had little of the market competition that privatization advocates point to as theprocess’s basic virtue. In fact, it was a relationship of both monopoly supplyand monopsony purchase, yet it worked well.

At the state and local levels of government in the United States, for example, governments have contracted out the contracting out of social service provision. For example, in Philadelphia, the social service system wasa complex maze of government employees and private contractors. The city spent US$612 million to provide services to 20,000 children, with 820social workers managing the services through a network of 348 contracts.The quality of services depended ultimately on the ability of the public andprivate nodes of the network to connect. The New York City childcare systemhas likewise been plagued with the difficulty of managing hundreds of agencies to which the city outsourced services (Young 1974).

The lesson is that the expansion of government contracting into more areashas not necessarily reduced government’s role. It has transformed its rolefrom supplier to manager of its contracts and has increased the need for it to be alert to the risks of monopolistic supply. This is not necessarily amore difficult job than managing a traditional authority-based hierarchy.Overseeing the contracting out of cafeteria or custodial services, for example,is relatively straightforward (although ensuring high-quality services througha reliable workforce is always a tough job). However, managing a con-tract requires a fundamentally different skill set – negotiating a contract, overseeing its execution, and assessing its results – from that required for managing direct service provision by government employees. As we shall see,

some contracted-out services are indeed more difficult to administer, but contracting out is not inherently more difficult to manage than direct serviceprovision. It is, however, fundamentally different, and those challengesescalate when government must work with monopolies.

Service integration

The complexities of these systems have increasingly led governments to relyon service integration contractors. Instead of purchasing goods and servicesdirectly from suppliers, government contracts with a general contractor, who in turn contracts out with other suppliers for the production of the goodor service. In many cases, the service integrator is responsible not only for general oversight of the project but also for designing many of its keyelements (Kettl 2009). Many new systems, from space shuttles to militaryfighting vehicles, have become so complex that government finds it easier tocontract out the entire system, from design through all of the productionstages. United Space Alliance is an example of such a service integration con-tract, but it extends much farther into functions as broad as social servicesand defense projects.

Such service integration contracts stretch to other levels of governmentand into other functional areas. The US Marine Corps, for example, investeda decade’s work and $1.7 billion into a new amphibious fighting vehicle.However, the system proved extremely trouble-prone. It broke down onceevery four and a half hours, leaked, and was hard to steer. The vehicle fellwell short of the Marine Corps’ expectations; but the contractor, GeneralDynamics, nevertheless received $80 million in bonuses for its work. The coreof the problem, government investigators concluded, was that the Pentagonhad insisted on an overly complex system and had contracted out much ofthe design and production work – it had contracted out the contracting out,and the Marines lacked sufficient expertise to oversee the process. Militaryplanners feared that they might have to start over with a fresh design and anew system (US GAO 2008b).

The rise of these service-integration contracts has had important implica-tions for the service system. As government has contracted out not only the production but also the design, intellectual capital has flowed to the managing contractors. The capacity of government agencies to understandfully what they are buying, how much the goods and services ought to cost, how well programs are working, and how to fix any problems erodesas expertise drifts to the private sector. NASA investigations into both the Challenger and Columbia accidents found that the agency had becomeheavily dependent on its contractors for life-and-death judgments. In theChallenger disaster, it was a private contractor that had the importantexpertise on the effect of cold weather on the rubber O-rings, the failure of

which doomed the shuttle seconds after launch. In the case of Columbia, NASAhad to rely on its contractors to assess the flightworthiness of the shuttle. It did not have sufficient expertise in-house to exercise independent final judgment (NASA 2003).

In all these cases, government employees played the role of contract manager. There was nothing inherently wrong with this role. The problemwas that its employees typically were trained as functional specialists. In the Marines, contract managers supervised the paper flow but lacked the expertise to make and oversee the key technical decisions. In Philadelphia,the employees charged with overseeing the contracts were social workers, who went to work for DHS to do social work. As the private and non-profit service network expanded, they became contract managers. Even thatcould have worked – except that the workers were trained as social workers,they had not been trained as contract managers, and the mismatch provedfatal to a 14-year-old girl, Danieal Kelly, whose problems the system failedto solve.

This unfortunate tale surfaces a broader set of problems. Not only is contracting out increasingly problematic because of the strategic mismatchbetween its ideological rationale – rely on the presumed superiority of private-market competition for delivering public services – and its reality – contractingout in many areas with substantial market imperfections; there is also thetactical mismatch in its management, between government employees hiredfor their substantial knowledge, in understanding programmatic goals, andthe functions these employees actually perform, in managing the contractorswho work to achieve those goals.

Contracting out has thus drifted a long way from the original goals thatdrove its growth. Market competition might indeed produce high-quality goods at low cost; but in many areas in government contracting, especiallyin the most important growth areas, there is little real market competition.In the process, much of government’s substantive work has drifted into the hands of government contractors, leaving government employees with con-tract management responsibilities for which they are poorly prepared. Theproblem lies not with the contracting-out strategy but with its execution – and with the faulty assumption that the magic of the markets will assurehigh-quality services at low costs. The result, too often, is like baking a cakewithout flour.

Public management strategies

Does this mean that the dramatic expansion of contracting out, both in theUnited States and in many other nations around the world, is a fundamentalmistake in the search for democratic governance? Do the performance problems that flow from this double mismatch argue for retreating on thecontracting-out movement?

Dependence on contractors

The first answer is that this is a hypothetical question. Governments havebecome so dependent on contractors that retreating on the strategy woulddemand a very large increase in the number of government employees – adecision that would be difficult indeed to support in a political era focusedon limiting government’s size and reach. Re-publicization is rare. For example, in the aftermath of the 11 September terrorist attacks, the federalgovernment decided to shift responsibility for airline security screening from the private sector to the government. It took a major terrorist attackto prompt such a decision, and the move stands in stark contrast to the deci-sion by many other nations to contract out security screening. In virtuallyall other cases, contracting out is expanding into new areas, in part becauseit sometimes is cheaper. It often allows governments to reduce the numberof public employees, and thus to create at least the illusion of smaller government (Light 1999). Most important, it allows government to obtainadditional operating flexibility and invaluable expertise that would be verydifficult to build within government itself. Rolling back the contracting-outmovement seems very unlikely.

Performance issues

Second, the performance problems that have come from the outsourcing movement are not necessarily inherent. Government cannot typically rely on self-regulating market competition to ensure high success and low cost,but it can strengthen its capacity to act as a “smart buyer”. It is one thingto contract out the production of goods and services to private suppliers. Itis quite another to contract out the basic decisions about what goods andservices to buy, how those goods and services should be designed, and howwell they work. No private automobile company would willingly surrenderto another firm the basic questions about what its cars ought to look like,how the sound systems would perform, or what mileage the engine shouldget; yet in some cases governments have contracted out just such basic deci-sions. It is scarcely surprising that such strategies have developed problems,for they bring neither the discipline of the markets nor the authority of governmental expertise. It will be impossible in most cases to restore mar-ket competition to those areas where it is lacking. Government would notwish to encourage private buyers for some goods, like nuclear weapons, andin many areas private sellers would not want to enter the markets with-out some assurance from government that their investments would produceprofits. Government faces the inescapable challenge of building the capa-city to manage its contracts more effectively.

Human capital issues

Third, this implies that a central core of government’s challenge in contractmanagement is a human resources issue. If government wants to ensure better performance in its contract systems, it will need to create and trainbetter governmental contract managers. This, of course, is an inherent prob-lem. In most governments, contract management is not a high-prestige field,and few government employees enter the public workforce with contract management as a career goal. Social workers join government to do socialwork; environmentalists to work on the environment; space scientists to dospace science. This creates the human resources dilemma of a governmentthat relies more on contracting out: those who join the government to pursue their substantive interests often end up in jobs that require them tomanage contracts instead, where others do the work for which they trained;and those with the skills to perform contract management find themselvesin lower-prestige positions. The effective management of contracts requiresgovernment human-resource planners to focus more attention on contract-ing out and to ensure that the employees overseeing contracts have the skillsneeded to do so. If social workers end up overseeing contracts without thenecessary competencies, they will be unhappy, and program management will surely suffer.

What competencies do contract managers need? They must have strongsubstantive backgrounds, so that they can understand the issues that bubbleup from the contractors; and they must be quick students in complex areas,so that they can steer contracts in the direction that governmental policyrequires. They must be adept negotiators. They must be good financial managers, for control of the money flow is typically the strongest control ofprogrammatic results. They must be good auditors, to ensure that moneyends up going where they intend. They must be able to evaluate the out-comes of these complex relationships among government, contractors andsubcontractors, so that they can assess whether government is getting its money’s worth. They must, in short, be experts in steering complex processes(US GAO 2009). These competencies, of course, are a long distance fromthe arguments, often ideological, that promote the spread of contracting out;but without these skills government is likely to be steered by the contractorsinstead of the other way around. Government does not need to know morethan its contractors. If it did, it might well not need them (or, at least, notneed as many of them). But it must know enough to ensure that the private contractors are held accountable to public policy.

Information issues

Fourth, government needs to build a far more effective information systemfor tracking how the money flows and what the contracts produce. In most

countries, it is impossible for anyone – even top government officials – toknow how much money contractors are spending, and what their role is in shaping each program. A product of the anti-government privatizationmovement is that its promoters believed at the core of their being that private provision of goods and services is so inherently superior and self-regulating that there was no need to manage or measure the process.Government, however, does indeed need to track where its money goes andwhat it buys. Sometimes the battle is over the basic assumption that govern-ment ought to behave more like the private sector. But no private companywould give its suppliers a blank check and fail to track how the money wasspent (Hatry 2007; Poister 2003). If government is going to rely more on theprivate sector and to be called on to act more like the private sector, thenit needs to follow other parts of this mantra as well, including building itscapacity to oversee the flow of its money (that is, of taxpayer’s money).

Transparency

Fifth, the contracting-out process needs far more transparency. A side-effectof the contracting-out movement has been an increasing difficulty in deter-mining just who is doing what on behalf of the people while spending thepeople’s money. The rise of service-integration contractors has even furthermuddied the trail, for it has made it even more difficult to determine who is making the important decisions. Moreover, contracting out is, of course,more than just about accomplishing the public’s business. For contractors,it is about getting, keeping and expanding business, and the contracting-outprocess has, not surprisingly, tended to create problems of corruption. Forexample, a former US Department of Defense official was indicted for steering the US$23.5 billion aerial tanker project to Boeing and then takinga job at the big aerospace contractor.

As more decision-making power has flowed to private-sector companies,the challenge of identifying, preserving and promoting the public interest hasgrown. Laws can prohibit corruption and reduce the chances of the appear-ance of corruption, but the conflicting incentives of a governance system withmore contracting out requires an even higher standard of transparency, some-times beyond the level that many private companies find comfortable.

This transparency can come in many ways. In the United States, for example, the federal government has created a database, USASpending.gov,that allows citizens to search for who gets how much money and where it goes. While a US Senator, Barack Obama was a prime architect of thewebsite, and his work on the project gave powerful clues about his instinctsfor web-based, information-driven accountability strategies. At the federallevel, the Government Accountability Office’s “high-risk list” has focused itsauditing and evaluation work on the programs most prone to fraud, wasteand abuse. But these efforts have often fallen short of public expectations

and have failed to direct public attention to the core of many problems. Moretransparency cannot solve these problems, but without it there can be nosolution at all.

Governance challenges

The deep realities of contracting out are very different from the rhetoric thathelped feed its growth and the public’s expectations about how it works. Whileoften portrayed as a new movement, it is in fact as old as government itself.While often argued as a way to reduce government’s size and scope, it hasoften brought government more deeply into the fabric of private organiza-tions. While often presumed superior to government action because of thepower of private markets, it often operates in arenas with substantial marketfailure. While often thought to be self-executing, it requires a strong and transformed role for government.

This does not make it good or bad in itself – just different from the waysin which government has traditionally been structured and governed. Indeed,contracting out is neither good nor bad. Governments have embraced it becauseof important, probably irreversible political and pragmatic realities. Govern-ment’s most fundamental governance challenge, therefore, is to discover moreeffective strategies to ensure it works well. There is an endless stream of talesof mismanagement that has come from poorly managed contracts. These tales will spread and worsen if the trend toward increasing government con-tracting grows without a simultaneous growth in its ability to manage andgovern the system it has created.

The job is eminently doable, but doing it will require a fresh approach togovernance – one that develops new approaches to govern the new realitiesof government contracting. Government will need to focus on performance,on the results that contracts produce, instead of on the processes along theway. It will need to develop human capital, especially inside public agencies,so that government has the capacity to oversee the contracts it negotiates.It will need effective information systems, reinforced by greater trans-parency, to provide more effective feedback. In short, it needs a governmentrobust enough to govern the systems it creates. It needs twenty-first-centuryapproaches to govern effectively the realities of its twenty-first-century governmental tools.

References

Albalate, Daniel, Bel, Germa and Fageda, Xavier (2007) Privatization and Regula-tion of Toll Motorways in Europe, University of Barcelona, Research Institute ofApplied Economics, IREA Working Papers.

Boston, Jonathan (2000) “Organizing for Service Delivery: Criteria and Opportunities,”in B. Guy Peters and Donald J. Savoie (eds) Governance in the Twenty-first

Century: Revitalizing the Public Service, Montreal: Canadian Centre for Manage-ment Development.

Brown, Lawrence D. and Jacobs, Lawrence R. (2008) The Private Abuse of the PublicInterest: Market Myths and Policy Muddles, Chicago, Ill.: University of ChicagoPress.

Donahue, John D. (1989) The Privatization Decision: Public Ends, Private Means,New York: Basic Books.

English, Linda M. and Guthrie, James (2003) “Driving Privately Financed Projectsin Australia: What Makes Them Tick?”, Accounting, Auditing and AccountabilityJournal, 16 (3): 493–511.

Evans, Andrew W. (2007) “Rail Safety and Rail Privatisation in Britain”, AccidentAnalysis and Prevention, 39 (3): 510–23.

Gellhorn, Walter, Byse, Clark, Strauss, Peter L., Rakoff, Todd and Farina, Cynthia R.(2003) Administrative Law, Cases and Comments, 10th edn, Mineola, NY: Founda-tion Press.

Hatry, Harry (2007) Performance Measurement, 2nd edn, Washington, DC: The UrbanInstitute.

Kettl, Donald F. (1993) Sharing Power: Public Governance and Private Markets,Washington, DC: The Brookings Institution.

Kettl, Donald F. (2002) “Managing Indirect Government”, in Lester M. Salamon(ed.) The Tools of Government: A Public Management Handbook for the Era of Third-party Government, New York: Oxford University Press.

Kettl, Donald F. (2009) The Next Government of the United States: Why OurInstitutions Fail Us and How to Fix Them, New York: W. W. Norton.

Light, Paul C. (1999) The True Size of Government, Washington, DC: The BrookingsInstitution.

McLaughlin, Kate, Osborne, Stephen P. and Ferlie, Ewan (2002) New PublicManagement: Current Trends and Future Prospects, Abingdon: Routledge.

National Aeronautics and Space Administration (NASA) (2003) Report of theColumbia Accident Investigation Board, at http://www.nasa.gov/columbia/caib/PDFS/VOL1/PART02.PDF

Osborne, Stephen P. (ed.) (2002) Public Management: Critical Perspectives, Abingdon:Routledge.

Padovani, Emanuele and Young, David (2006) “Managing High-risk Outsourcing”,Public Management, 88 (1): 29–32.

Padovani, Emanuele and Young, David (2008) “Toward a Framework for ManagingHigh-risk Government Outsourcing: Field Research in Three Italian Muni-cipalities”, Journal of Public Procurement, 8 (2): 215–47.

Poister, Theodore H. (2003) Measuring Performance in Public and NonprofitOrganizations, San Francisco, Calif.: Jossey-Bass.

Pollitt, Christopher and Bouckaert, Geert (2000) Public Management Reform: AComparative Analysis, Oxford: Oxford University Press.

Preston, John (1996) “The Economics of British Rail Privatization: An Assessment”,Transport Reviews, 16 (1): 1–21.

Salamon, Lester M. (ed.) (2002) The Tools of Government: A Public Management Hand-book for the Era of Third-party Government, New York: Oxford University Press.

Savas, E. S. (2000) Privatization and Public–private Partnerships, New York: ChathamHouse.

Schick, Allen (1996) The Spirit of Reform: Managing the New Zealand State Sectorin a Time of Change, Wellington: State Services Commission.

United States Government Accountability Office (2008a) “GAO’s High-risk Areas asof March 2008”, at http://www.gao.gov/docsearch/featured/highrisk_march2008.pdf

United States Government Accountability Office (2008b) Defense Acquisition:Assessments of Selected Weapon Programs, GAO-08-467SP, Washington, DC: USGovernment Accountability Office, at http://www.gao.gov/new.items /d08467sp.pdf

United States Government Accountability Office (2009) Human Capital: Oppor-tunities Exist to Build on Recent Progress to Strengthen DOD’s Civilian Human CapitalStrategic Plan, GAO-09-235, Washington, DC: US Government AccountabilityOffice, at http://www.gao.gov/products/GAO-09-235

Young, David W. (1974) “Referral and Placement in Child Care: The New York CityPurchase of Service System”, Public Policy, 22 (3): 293–327.

GOVERNANCE OF OUTSOURCING AND

CONTRACTUAL RELATIONSHIPS1

Federica Farneti, Emanuele Padovani and David W. Young

Many public sector organizations (PSOs) rely on outsourcing as a way of increasing their “value for money” in the provision of public services(Broadbent and Guthrie 2008). As a result, the topic of outsourcing has been addressed with some regularity in the public management literature formany years. Increasingly, attention is being directed toward the nature ofthe risk associated with an outsourced activity, and the resulting implicationsfor (Considine 1999, 2001; Considine and Lewis 2003) PSO managers.

In this chapter, we present a framework that PSO managers can use to assess the nature of the risk they face in an outsourcing decision, and juxtapose it with alternative governance models. We argue that, as the riskassociated with an outsourced service increases, so, too, does the sophistica-tion of the governance model a PSO needs to use to manage it. Thus, thedecision to outsource a high-risk service must depend, in part, on whetherthe PSO has the capability to use the requisite governance model.

We begin by discussing the roots of outsourcing, with a particular focuson Four-governance Model framework. We assess his framework in light ofour empirical research in Italian PSOs that led to our model for assessingoutsourcing risk, and we argue that a PSO does not need to have a singlegovernance model, or even a highly sophisticated one. Rather, it may needseveral governance models operating simultaneously, each attuned to the riskassociated with the outsourced service being managed.

The outsourcing context

In some countries, the impetus for outsourcing has been legislative. In the UK,for example, although competitive outsourcing had been taking place for sometime, the Local Government Act of 1988 required it at the municipal level.Similarly, as part of its public-sector reform initiative, New Zealand required

municipalities to use public–private competition to improve local public services. In Australia, where there has been rapid growth in public–privatepartnerships (PPPs), some states, such as Victoria, require that half of alllocal public service expenditures be submitted to competitive bidding. As a result, many public services are now provided not only by PSOs, but byprivate-sector firms as well (Broadbent and Laughlin 2003; English and Guthrie2003). In part the shift has been due to the commonly acknowledged factthat private-sector organizations are almost always more efficient in service-provision than public-sector ones (Kulmala et al. 2006).

However, a move toward outsourcing does not need a legislative mandate.In Italy, for example, although local government reforms encouraged out-sourcing, they did not require it; and yet, by 1999, between 56 and 93 percentof the most important local services (natural gas distribution, waste disposaland collection, water and sewerage) were being outsourced (Enea-Nomisma1999). In the US, although there are no legislative mandates, outsourcinghas been going on for some time (Anthony and Young 2003). In particular,it began to grow rapidly in the 1980s under the political pressure of PresidentReagan’s “New Federalism”.

Outsourcing and contemporary public management

Outsourcing is a logical consequence of efforts by PSOs over the last threedecades to minimize waste and to promote efficiency and effectiveness (Nolanand Brendan 2001). It has been fostered in many PSOs by a shift towardgreater managerial autonomy, a move toward private-sector managementapproaches, and the promotion of competition among private-sector firmswith the dual goals of decreasing costs and increasing quality.

Outsourcing also has been driven by a PSO’s desire to increase its flexi-bility (Hartmann and Patrickson 2000) as well as to transfer some of its riskin running a program to private entities, while simultaneously allowing it to maintain control over the program’s essential elements (Quiggin 1996).And, finally, outsourcing has allowed a PSO to draw on expertise not available in-house (Young 2000), thereby permitting it to use private-sectorinnovation, technology and expertise (Ponomariov and Kingsley 2008). Ineffect, many PSOs have moved quite deliberately and systematically fromthe left side of Figure 15.1 to the right side. Moreover, in conjunction withits shift to PPPs, the state of Victoria (Australia) also began to outsourcesome activities, such as human services, that previously were considered tobe in the exclusive domain of the public sector (Teicher et al. 2006). Similarshifts have taken place in other countries as well (Tremblay et al. 2008).

With its focus on the private sector, outsourcing has also become a keyelement of the New Public Management paradigm (Hood 1991, 1995) and,more recently, the New Public Governance model (Box et al. 2001; Pollitt 2002; Cabrero 2005; Osborne 2006). It is being used in efforts to transform

of Considine’s four models (procedural, corporate, market and network) has a distinct source of rationality, form of control, primary virtue, and servicedelivery focus.

Model 1: procedural governance

The procedural governance model (PGM) is characterized by centralized andhierarchical public administration, and is rule-bound, with protocols anddefined practices. There is a strong top-down authority that requires layersof checking to reduce the likelihood of errors. The PGM was developed at atime when most PSOs were engaged in minimal outsourcing, and is designedto guarantee that standardized services are provided at the lowest per-unit cost.

Model 2: corporate governance

The corporate governance model (CGM) emerged because of a perceptionthat the PGM did not fit with a variety of administrative requirements forsuccessful outsourcing, mainly the need to maintain greater control over pub-lic expenditures (Pallot 1992; Pierre and Peters 2000). In addition, the PGMwas not able to deal with the increasing complexity of government (Lapsley1988) and the need to target some services for a subset of the citizenry.

In the CGM, considerable importance is given to planning, budgeting and reporting. A PSO using the CGM concentrates on outputs instead ofon inputs, and it focuses on specific groups of citizens who are receiving services. In effect, its emphasis shifts from adhering to arbitrary rules to achiev-ing results.

Model 3: market governance

In the market governance model (MGM), a PSO encourages competitionamong potential vendors, and develops contracts to define its relationshipswith the selected ones (English et al. 2005). These contracts stress quality aswell as cost, and place considerable emphasis on meeting citizen needs(Pierre and Peters 2000).

The MGM may include arrangements with a commercial company, a publicauthority, a nonprofit organization, or all three (Goldsmith and Eggers 2004;Pollitt 2003). Its use of performance-based service contracting encouragesvendors to be innovative and seek increasingly cost-effective ways to deliverservices (United States General Accounting Office 2002).

In Australia, where the MGM began in the mid-1990s, PSOs at variouslevels of government have introduced “best commercial practices”. In someinstances, a PSO has used the “corporate form” for its business activities,and in others it has “privatized” a governmental unit by selling the relevantassets to an outside entity (Broadbent and Guthrie 2008).

By creating market dynamism and increased autonomy within the publiccontext, the MGM can help to ensure accountable managerial behavior.However, for many PSOs, the MGM can also present some difficult man-agerial challenges (Osborne and Gaebler 1992). This is because, under theMGM, a PSO needs to be certain that it measures the vendor’s outputs (ratherthan inputs), and structures an appropriate set of reporting and feedbackrelationships. For many PSOs, developing the requisite management-controlsystem runs contrary to their longstanding, input-based managerial cultures,and therefore is a difficult transition to make (Padovani and Young 2008).

Model 4: network governance

Although the MGM strives to increase competition so as to help containcosts, its contracts generally focus on only one service. It therefore is inappro-priate when some outsourced services need to be coordinated with others.

The network governance model (NGM) is designed to overcome this limitation. It does so, in part, by focusing on linkages among vendors; and,in so doing, responds to complex citizen needs. Indeed, because of theimportance of managing the linkages among these needs, some researchershave argued that PSOs must move toward the NGM model if they are toserve their citizenry most effectively (Mandell 2001).

The goal of the NGM is to combine a high level of public–private collaboration (characteristic of the MGM) with a robust network of serviceproviders. To achieve this goal, PSOs first establish long-term relationshipswith their vendors. They then use sophisticated information technology toconnect the entire network of vendors (and other involved organizations) soas to give citizens a wide range of service-delivery options. Under the NGM,a key task of government is the management of these networks (McGuire2002; McLaughlin and Osborne 2005; Klijn 2008).

The essence of the NGM, then, is cooperation among government, non-profit and (sometimes) private-sector organizations to help ensure a high levelof service coordination (Considine and Lewis 2003). In so doing, the NGMhas spawned a stream of research that examines its impact on enhancing acommunity’s social capital (Moran 2005). It also is conceivable that the NGMcould influence a community’s relational capital – a topic that has been exploredconsiderably in the private sector (Sawhney and Zabin 2001) and has clearapplicability to the public sector.

Relational capital is important if a PSO is successfully to outsource a complex set of activities, such as social services or pollution control, whereit must develop partnership-like coordination among a wide network of vendors. These relationships can include “joint investments, shared research,common development ventures, and flexible methods for linking financiers,regulators and a host of public and private service providers” (Considine 2001: 30). They can also include the sharing of cost information with other

GOVERNANCE OF OUTSOURCING

public or private organizations in an effort to meet a complex set of citizenneeds as cost-effectively as possible.

As with the MGM, many PSOs are not prepared for the challenges posed by the NGM. Because managing their networks frequently requiresskills, knowledge and technology that they do not possess, PSOs have hadvarying degrees of success in attempting to use the NGM (Sørensen 2002;Goldsmith and Eggers 2004).

Moving from the PGM to the NGM

In all four governance models, PSOs are attempting to improve the qualityand cost-effectiveness of their public services. However, movement from thePGM to the NGM entails developing new skills and implementing new tech-nology so as to manage vendors differently. The result is an increase in thecost of governance as a PSO moves along the spectrum toward the NGM.

Given the possible higher costs of the more sophisticated governance models,the central question is not whether a PSO should be attempting to move asdeliberately as possible from the PGM to the NGM, but rather which of thefour governance models is appropriate for each service being outsourced. Thus,it is conceivable that a PSO could have several governance models operat-ing simultaneously, each attuned to the nature of the service being outsourced.

A strategic framework for outsourcing

Before deciding on a governance model, a PSO must decide whether it willoutsource a particular service or provide it itself. Part of this decision entailsassessing the risk of low-quality and/or unduly expensive vendor performance,and considering what that might mean for the overall satisfaction of the citi-zenry. For example, if a city or town decides to have a youth program brochureprepared by an outside vendor, the citizenry most likely is unaware of, andunaffected in any significant way by, the vendor’s performance.

By contrast, when a vendor provides a service directly to the citizenry, ratherthan to the PSO itself, there is a considerable potential for citizen dissatis-faction. Waste collection, snow removal and street repair are all examples ofservices where the citizenry is directly affected by the choices, able to assess theirquality (however subjectively), and concerned about the resulting cost. Inthese latter instances, despite its potential for enhancing value to the citizenry,outsourcing can present difficulties in managing the provision of public services.

The need for risk assessment

Because of these differences in risk, a PSO that is considering outsour-cing an activity must assess the risk associated with unsatisfactory vendor performance. In the case of printing a brochure, for example, the risk is low.

Clearly, quality specifications can be included in the contract, and the resultscan be easily monitored. But this is not the central issue. Rather, the questionis: What are the consequences of poor vendor performance? In the brochuredecision, the answer is that it will have little impact on the citizenry, and thevendor can be replaced easily, if necessary, for future printings.

On the other hand, services such as waste collection, water and sewerageprovision, and traffic control have quality and service goals that are moredifficult to measure and monitor. Moreover, poor vendor performance willdirectly affect the citizenry, and the PSO may have difficulty replacing a poorlyperforming vendor in a timely way. Clearly, the risk is higher.

A great deal of literature has emerged in recent years concerning out-sourcing risk in both the public and the private sectors. In some instances,risk has been largely of a financial nature, and has been assessed in light of the difficulty of providing services within a set of budgetary constraints(Ball et al. 2003). In others, risk has been viewed in terms of a vendor’s capacity to deliver the requisite services, or the uncertainty of the task to be performed (Venkateswar 2005). In still others, the focus has been on thelogistics of managing the vendor (Steane and Walker 2000). In all of theseareas, a central issue has been the information asymmetry between the vendor and the PSO, which can lead the PSO to assume more risk than itthinks it is (Demsetz 1968).

From the perspective of a PSO’s outsourcing decisions, risk can beaddressed more strategically than these uni-dimensional approaches. To do so, a PSO must focus on such matters as: (a) increased dependence on external suppliers, resulting in a potential loss of control over critical activities; (b) greater difficulty in cost management, especially when there are adversarial relationships; (c) loss of either essential competencies in thepublic entity or control over the resources that it would need to acquire inorder to resume conducting the activity itself; and (d ) loss of the flexibilityto respond to changing needs of the citizenry (Kettl 1993; Quinn and Hilmer1994; Domberger 1998).

Most of these latter concerns relate to a three-way juxtaposition of (a) the impact of the service on the citizenry, (b) the competitive nature ofthe market for vendors, and (c) the ease or difficulty of switching vendors (or returning to internal service provision) if a given vendor’s performanceis unsatisfactory (Padovani and Young 2006, 2008). With the preparation ofa brochure, for example, there is a low potential for citizen dissatisfaction,the market is highly competitive, and the PSO would find it quite easy toswitch from one vendor to another. By contrast, not only is a service suchas waste removal highly important to the citizenry, but also there may befew other capable vendors in the marketplace, such that the PSO would havedifficulty finding a replacement if the vendor’s performance were unsatisfactory.Moreover, even if a replacement could be found, the cost of switching vendors might be high.

be measured by complaints, which typically were generated by tourists andwere dealt with by the municipality’s contract manager or its elected officials.Over the period of the contract there was a growth in complaints and con-siderable citizen dissatisfaction.

In this instance, the municipality’s use of the procedural governancemodel was inappropriate. It resulted in a focus on regulating the vendor’sactivities, rather than on measuring its effectiveness in meeting citizen needs.It seems clear that either the corporate governance or the market govern-ance model, with their emphasis on quality and citizen needs, would havebeen more appropriate.

Beta Municipality

Beta Municipality was a residential and industrial town of 30,000 inhabit-ants, also in northeast Italy but located inside a large metropolitan area. Itfaced many of the same issues as Alpha Municipality. Here, however, thecontract with the vendor was results-oriented – the focus was on output targets to be accomplished by the vendor instead of on specific activities. Ona daily basis, the contract manager inspected such output indicators ascleanliness of the streets, overflowing containers, and facility maintenance.Negative inspection-findings (as well as citizen complaints) were communi-cated to the vendor daily, and the vendor was required to notify the contractmanager when each problem was resolved.

Structural deficiencies – instances when the same problems occurredrepeatedly and needed to be solved more systemically – were discussed dur-ing bi-monthly meetings of a joint commission (composed of the contractmanager, a vendor representative, and a citizen representative). The commis-sion also levied penalties on the vendor for noncompliance.

As part of its MGM, Beta Municipality had a management control system that assisted the town’s elected officials and senior managers in monitoring two very important performance indicators: percentage of recyclables and purity of recyclables. All other output indicators – such asstreet cleanliness, availability of solid-waste and recyclable containers, andthe quality of facility maintenance – were delegated to the contract manager.With this governance model, there was considerable cooperation between the municipality and the vendor to meet citizen needs; and, as a result, thenumber of citizen complaints was minimal.

Gamma Agency

Gamma Agency was a public transportation entity that had outsourced themaintenance of its copy-center service – an internal service that provided flyers, timetables and other similar documents for citizens. The contract wasprocedural, with detailed specifications of activities to be delivered within

specific deadlines (e.g. maintenance of specific parts of the machines atdefinite dates or after a specified number of copies had been printed) or onrequest (e.g. substitution of toners, repairing of damaged parts, etc.). In effect,Gamma Agency was using the PGM for a service where adherence to pro-cedures was important.

Case comparisons

These three cases illustrate the importance of attaining a fit between the ser-vice being outsourced and the governance model used to manage the vendor.For example, because it employed the PGM model, Alpha Municipality wasnot able to address adequately the problems that arose with poor vendorperformance. By contrast, in Beta Municipality, where the level of risk wasthe same as that of Alpha Municipality, use of the MGM helped to ensurethat the service met the citizens’ needs. On the other hand, Gamma Agency’suse of the PGM was quite appropriate under the circumstances.

Conclusions

These three cases illustrate several important conceptual points. First, underthe right circumstances, a simple governance model such as the PGM maybe all that is needed to ensure adequate vendor performance.

Second, the use of an inappropriately simple governance model for a high-risk service can lead to problems with vendor performance, and thus can resultin citizen dissatisfaction.

Third, from these case examples, one can conclude that choosing the most appropriate governance model is not a one-time event. If, for example,a PSO modifies its contract with a vendor to include a more complex mixof services, thereby creating an increase in its risk, it needs to assess the fit of its existing governance model with the new needs. Thus, as a PSO grows,as its outsourced services increase in complexity, and as the associated riskincreases, it needs to shift to increasingly sophisticated governance models.

Fourth, before it decides to outsource a high-risk service, a PSO’s seniormanagement needs to be certain that it has the capability to use the requisite governance model; and, if that model is other than the PGM, the municipality needs a management control system that is sufficiently sophisticated to manage the vendor’s activities.

Fifth, a PSO’s management control system is an important element of theCGM, the MGM and the NGM. In Beta Municipality, for example, it wasa key element in the effort to ensure adequate vendor performance.

Finally, given the above, it seems quite obvious that a PSO can have multiple governance models operating simultaneously. In part, the appro-priate governance model for a given outsourced service depends on the degreeof risk. Using the NGM for a low-risk service, for example, is excessively

costly for the task at hand. At the same time, the decision to outsource a high-risk service must depend, in part, on whether the PSO has the capability tooperate the requisite governance model. For example, to outsource a servicethat requires a high degree of coordination, but without possessing the man-agerial and technological capability for the NGM, may lead to redundant,conflicting and ineffective services. In the first instance, the cost of governanceis excessive. In the second, insufficient (or inappropriate) resources are beingdevoted to the coordination effort, resulting in a potentially dissatisfied citi-zenry. Neither should be acceptable to a PSO’s senior management team.

Note1 This chapter expands upon and develops the argument in F. Farneti and D. Young

(2008) “A Contingency Approach to Managing Outsourcing Risk in Municipalities”,Public Management Review, 10 (1): 89–100.

ReferencesAnthony, R. N. and Young, D. W. (2003) Management Control in Nonprofit Organ-

izations, 7th edn, Burr Ridge, Ill.: McGraw-Hill/Irwin.Ball, R., Heafey, M. and King, D. (2003) “Risk Transfer and Value for Money in

PFI Projects”, Public Management Review 5 (2): 279–290.Box, R., Marshall, G., Reed, B. J. and Reed, C. (2001) “New Public Management

and Faux Democracy, and the Challenge to Democratic Governance”, PublicAdministration Review, 61 (5): 608–19.

Broadbent, J. and Guthrie, J. (2008) “Public Sector to Public Services: 20 years of ‘Alternative’ Accounting Research”, Accounting, Auditing and AccountabilityJournal, 21 (2): 129–69.

Broadbent, J. and Laughlin, R. (2003) “Public Private Partnerships: An Introduc-tion”, Accounting, Auditing and Accountability Journal, 16 (3): 332–41.

Cabrero, E. (2005) “Between Public Management and New Public Governance: The Caseof Mexican Municipalities”, International Public Management Review, 6 (1): 76–99.

Considine, M. (1999) “Markets and the New Welfare State: Employment AssistanceReforms in Australia”, Journal of Social Policy, 28 (2): 183–203.

Considine, M. (2001) Enterprising States: The Public Management of Welfare-to-work,Cambridge: Cambridge University Press.

Considine, M. and Lewis, J. M. (2003) “Bureaucracy, Network, or Enterprise?Comparing Models of Governance in Australia, Britain, the Netherlands, and NewZealand”, Public Administration Review, 63 (2): 131–40.

Demsetz, H. (1968) “Why Regulate Utilities?”, Journal of Law and Economics, 11:55–65.

Domberger, S. (1998) The Contracting Organization: A Strategic Guide to Outsourc-ing, New York: Oxford University Press.

Enea-Nomisma (1999) Rapporto Enea-Nomisma sui servizi pubblici locali, Rome: Enea-Nomisma.

English, L. and Guthrie, J. (2003) “Driving Privately Financed Projects in Australia:What Makes Them Tick?”, Accounting Auditing and Accountability Journal, 16 (3):493–511.

English, L., Guthrie, J. and Parker, L. (2005) “Recent Public-sector Financial Man-agement Change in Australia: Implementing the Market Model”, in J. Guthrie, C. Humphrey, O. Olson and L. Jones (eds) International Public Financial ManagementReform: Progress, Contradictions and Challenges, Greenwich, Conn.: InformationAge Press.

Farneti, F., and Young, D. W. (2008) “A Contingency Approach to ManagingOutsourcing Risk in Municipalities”, Public Management Review, 10 (1): 89–99.

Goldsmith, S. and Eggers, W. D. (2004) Governing by Network: The New Shape ofthe Public Sector, Washington, DC: The Brookings Institution.

Hartmann, L. and Patrickson, M. (2000) “Externalizing the Workforce: AustralianTrends and Issues for HRM”, International Journal of Manpower, 21 (1): 7–20.

Hood, C. (1991) “A Public Management for All Seasons?”, Public Administration,69 (1): 3–19.

Hood, C. (1995) “The ‘New Public Management’ in the 1980s: Variations on a Theme”,Accounting Organization and Society, 20 (2–3): 93–109.

Hutt, K. and Walcott, C. (1990) Governing Public Organizations: Politics, Structure,and Institutional Design, Pacific Grove, Calif.: Brooks/Cole Publishing Company.

Kettl, D. F. (1993) Sharing Power: Public Governance and Private Markets, Washing-ton, DC: The Brookings Institution.

Klijn, E. H. (2008) “Governance and Governance Networks in Europe”, PublicManagement Review, 10 (4): 505–25.

Koppenjan, J. and Klijn, E. (2004) Managing Uncertainties in Networks, London:Routledge.

Kulmala, H., Ojala, M., Ahoniemi, L. and Uusi-Rauva, E. (2006) “Unit CostBehavior in Public Sector Outsourcing”, International Journal of Public SectorManagement, 19 (2): 130–49.

Lapsley, I. (1988) “Research in Public Sector Accounting: An Appraisal”, Accounting,Auditing and Accountability Journal, 1 (1): 21–33.

McGuire, M. (2002) “Managing Networks: Propositions on What Managers Do and Why They Do It”, Public Administration Review, 62 (5): 599–609.

McLaughlin, K. and Osborne, S. P. (2005) “The New Public Management inContext”, in K. McLaughlin, S. P. Osborne and E. Ferlie (eds) New PublicManagement: Current Trends and Future Prospects, London: Routledge.

Mandell, M. (2001) “Collaboration through Network Structures for CommunityBuilding Efforts”, National Civic Review, 90 (3): 279–87.

Moran, P. (2005) “Structural and Relational Embeddedness: Social Capital andManagerial Performance”, Strategic Management Journal, 26 (12): 1129–51.

Nolan, Brendan C. (2001) “Introduction”, in Brendan C. Nolan (ed.) Public SectorReform: An International Perspective, Basingstoke: Palgrave: Macmillan.

Osborne, S. (2006) “The New Public Governance?”, Public Management Review, 8 (3): 377–87.

Osborne, D. and Gaebler, T. (1992) Reinventing Government: How the Entre-preneurial Spirit Is Transforming the Public Sector, Reading, Mass.: Addison-Wesley.

Padovani, E. (2005) Il governo dei servizi pubblici locali in outsourcing: Il controllodell’efficacia [The Governance of Outsourced Local Public Services: Controlling theEffectiveness], Milan: FrancoAngeli.

Padovani, E. and Young, D. W. (2006) “Managing High Risk Outsourcing”, PublicManagement, 88 (1): 29–33.

Padovani, E. and Young, D. W. (2008) “Toward a Framework for Managing High-risk Government Outsourcing: Field Research in Three Italian Municipalities”,Journal of Public Procurement, 8 (2): 215–47.

Pallot, J. (1992) “Elements of a Theoretical Framework for Public SectorAccounting”, Accounting Auditing and Accountability Journal, 5 (1): 38–59.

Peters, G. (1996) The Future of Governing: Four Emerging Models, Lawrence, Kan.:University Press of Kansas.

Pierre, J. (1999) “Models of Urban Governance: The Institutional Dimension of UrbanPolitics”, Urban Affairs Review, 34 (3): 372–96.

Pierre, J. and Peters, G. B. (2000) Governance, Politics and the State, London:Macmillan.

Pollitt, C. (2002) “The New Public Management in International Perspective: AnAnalysis of Impacts and Effects”, in Kate McLaughlin, Stephen P. Osborne andEwan Ferlie (eds), New Public Management: Current Trends and Future Prospects,London: Routledge.

Pollitt, C. (2003) The Essential Public Manager, Philadelphia, Pa: Open UniversityPress.

Ponomariov, B. and Kingsley, G. (2008) “Applicability of the Normative Model ofOutsourcing in the Public Sector: The Case of a State Transportation Agency”,Public Organization Review, 8: 253–72.

Quiggin, J. (1996) “Competitive Tendering and Contracting in the Australian PublicSector”, Australian Journal of Public Administration, 55 (3): 49–57.

Quinn, J. B. and Hilmer, F. G. (1994) “Strategic Outsourcing”, Sloan ManagementReview, 35 (4): 43–55.

Sawhney, M. and Zabin, J. (2001) The Seven Steps to Nirvana, New York: McGraw-Hill.

Sorensen, E. (2002) “Democratic Theory and Network Governance”, AdministrativeTheory and Praxis, 24 (1): 693–721.

Steane, P. D. and Walker, D. H. T. (2000) “Competitive Tendering and ContractingPublic Sector Services in Australia – a Facilities Management Issue”, Facilities, 18 (5–6): 245–55.

Teicher, J., Van Gramberg, B. and Holland, P. (2006) “Trade Union Responses to Outsourcing in a Neo-liberal Environment: A Case Study Analysis of theAustralian Public Sector”, Asia Pacific Business Review, 12 (2): 243–56.

Tremblay, M., Patry, M. and Lanoie, P. (2008) “Human Resources Outsourcing in Canadian Organizations: An Empirical Analysis of the Role of OrganizationalCharacteristics, Transaction Costs and Risks”, The International Journal of HumanResources Management, 19 (4): 683–715.

United States General Accounting Office (2002) Contract Management: Guidance Neededfor Using Performance-based Service Contracting, GAO-02-1049, Washington, DC:US General Accounting Office.

Venkateswar, N. R. (2005) “Mitigating Operational Risk in Outsourcing”, DMReview, 24 May 2005.

Young, S. (2000) “Outsourcing: Lessons from the Literature”, Labour and Industry,10 (3): 97–118.

THE GOVERNANCE OFCONTRACTING RELATIONSHIPS:“KILLING THE GOLDEN GOOSE”

A third-sector perspective

Steven Rathgeb Smith and Judith Smyth

Throughout the world, the role of third-sector organizations in providingpublic services is increasing, primarily through a sharp rise in governmentcontracting with these organizations.1 The result is profound changes in their governance, including the role of staff and boards, their programmaticfocus and their internal operations. Moreover, some changes are fundamentaland raise fears that through contracting governments may be “killing thegolden goose”. Contracting can also affect the motivation and character-istics of staff and board members and their accountability to stakeholders,including funders and service users. In many countries, the differencebetween traditional grants and contracts is declining as government com-missioners increase the specificity and performance expectations on their grants,creating new governance and accountability challenges for third-sectororganizations and their staff and volunteers.

The attraction of contracting with third-sector organizations – the “goldengoose” – reflects many factors: pressure to contain costs, new attention tocivil society organizations and their potential to build social capital and community, and the influence of the New Public Management, which seeksto improve the efficiency and effectiveness of public services through con-tracting, privatization and decentralization (see Hood 1991; Rhodes 1996;Considine 2003; Behn 2001; Kettl 2005). Given these factors, third-sector organizations offer at least the possibility of service innovation, lower costs,community and citizen engagement, voluntarism, and enhanced competitionwith the public sector. The community roots of many third-sector organ-izations also means that these organizations may have political support andinfluence, and represent politically important constituencies; thus, governmentcontracting with these organizations may be a strategy for government toreach out and support a particular group or constituency. Over time, though,

the potential benefits of contracting are often undermined by governmentpolicy, creating difficult choices and trade-offs for third-sector organizationsand the individuals who use their services.

This government reform movement has had far-reaching effects on third-sector organizations and the relationship between government and these organizations. It has fueled the growth of a wide variety of organizations to provide an array of public services, especially in health, social services and housing. Indirectly, the government reform movement has facilitated the increased interest in voluntarism, citizen participation and community service. Over time, the sharp rise in contracting has also encouraged third-sector organizations to become more assertive politically, either directlythrough their own actions or indirectly through associations representing theirinterests. In the UK and the US, these infrastructure organizations havereceived government funding in order to support the capacity of their third-sector members to compete for contracts and rise to the challenges of regulation and inspection.

In the US and the UK, the increasing role of third-sector organizationsin the last twenty years has also changed the role of government at all levels. Indeed, a key argument of this chapter is that the core principles ofgood public management – appropriate training and education of govern-ment managers, greater transparency and accountability, improved citizenengagement – drive all organizations that provide public services toward an increasingly standard offer regardless of the governance arrangements.These resulting effects are rarely raised or questioned whilst government andthird-sector agencies wrestle with the knotty governance dilemmas raised byan increasingly complex policy process and service delivery system createdby the growth of contracting.

An analysis of these governance challenges and dilemmas is the focus of this chapter, with an emphasis on the implications for public managersand third-sector staff and volunteers of the evolving contracting relationship.Thus, this chapter concentrates on third-sector organizations that receive substantial grants and contracts, with a particular emphasis on the impactof these contracts on the governance and programming of these organiza-tions. The research on contracting practices informing the analysis andfindings in this chapter is drawn primarily from the US and the UK in order to encourage international comparison and learning. These countriesshare similar trends in the overall growth of contracting with third-sectororganizations and the increasing emphasis on achieving good performancethrough contract compliance, regulation and inspection, with its attendanteffects on the governance of the third sector. Both countries illustrate moregeneral issues of governance pertaining to contracting that are also applicableto other countries and regions with very different contracting arrange-ments and regulatory regimes. However, public policy toward the third sector in the UK diverges from the US owing to the Compact (2009), which

THE GOVERNANCE OF CONTRACTING RELATIONSHIPS

is a formal structured process for the government and third-sector organiza-tions in the UK to work together. Over time, the Compact has encouragedgovernment to undertake other initiatives to support the third sector basedupon this Compact framework.

Throughout the chapter, the term “third-sector organization” will be used instead of nonprofit organization, NGO and voluntary organization.Third-sector organizations refer to the diverse set of organizations that can provide an array of services including health and social care, education,conservation and recreation, and advocacy. These organizations may be entirelyvoluntary or they may receive substantial public subsidies and contracts. Ingeneral, the term “nonprofit organization” is most widely used in the US,while “third-sector organization” has emerged as the most common term torefer to these organizations in Europe, Australia and New Zealand. “NGO”is most widely used in developing countries and can refer to a diverse mix of intermediary organizations. In the United States, the term “NGO”has generally referred to nonprofit organizations that have a non-US focus. In the UK, a similar distinction exists between voluntary organizations and NGOs (see Harris and Rochester 2001), and in the last decade “thirdsector” has been replacing “voluntary organization” as the favored descrip-tor of the sector – for example, the UK government has established an Officefor the Third Sector.

Importantly, the types of organization which regard themselves as part ofthe third sector are shifting. For example, the growth of hospices in the UKand the US creates new members of the third sector, but very few hospicesformally involve themselves in the third-sector community or in their rep-resentative bodies such as the Independent Sector in the US or the NationalCouncil of Voluntary Organisations (NCVO) in the UK. Housing associa-tions in the UK are in business and social terms very successful third-sectororganizations but nowadays rarely involve themselves with general third-sector infrastructure work. In the UK, new forms of National HealthService (NHS) hospitals and schools have been established. Although theyare not yet classified as third-sector organizations, they are in many waysconstitutionally and positionally much more like third-sector providers thantraditional public services. In the US, nonprofit hospitals are increasingly com-plicated mixed public and private entities, often with for-profit partnershipsand subsidiaries. These emergent and changing forms of “ownership” shouldperhaps be seen as part of the third sector. Likewise, new social-enterpriseorganizations such as leisure services companies are assuming responsibilityfor the delivery of a range of services previously provided by government inthe UK. For the most part, the traditional third sector has not embracedthese new organizations into their fold, despite the income potential frommembership fees locally and the potential to add to support for the sectornationally. Coming as they do from traditional public service governance,with very different psychological contracts and roots, these organizations

appear to be on different journeys to more traditional third-sector organ-izations such as social service agencies.

The chapter initially chronicles the growth of contracting in different countries, followed by a discussion and analysis of the key challenges anddevelopments affecting government and third-sector organizations involvedin contracting. The focus in this section of the paper is on three key issues:performance management and accountability; self-regulation, inspection,accreditation and quality frameworks; and competition. Each of these chal-lenges poses a threat to the “golden goose”, thus the section examines theimplications of these challenges for third-sector organizations, especially theirgovernance. The chapter concludes with suggestions for research, policy andpractice.

The growth of contracting

Government funding of third-sector organizations has a long history in many countries. In the United States, some of the oldest nonprofit organiza-tions in the country, such as Massachusetts General Hospital and HarvardUniversity, received government funding in the colonial period (Smith andLipsky 1993; Salamon 1987). In many European countries, including theNetherlands, the United Kingdom, Germany, Denmark and Sweden, volun-tary organizations emerged in the nineteenth century as crucial to providingimportant social and health services (Burger and Veldeer 2001; Lundstromand Svedberg 2003; Henriksen and Bundesen 2004; Bode forthcoming).However, the growth of the welfare state in the early and mid-twentieth century in countries such as Denmark, the UK and Sweden resulted in a long-term decline in the role of third-sector organizations in providing services(Lundstrom and Svedberg 2003; Henriksen and Bundesen 2004; Gutch1992). In these countries, local government tended to provide the bulk ofsocial and health services, with voluntary organizations providing specific niche services such as emergency assistance. In the US, the situation was quite different. Public social welfare services were provided primarily by localgovernment and were quite limited in nature. And the voluntary sector was quite small and narrow in scope. In the 1960s, government contractinggrew sharply with the growth of the national government’s role in healthand social policy.

Subsequently, many governments shifted their initial funding of third-sector organizations to more formal contracts with competitive tendering and substantial regulation. This change, which began in the 1970s and thenquickened in the 1990s and thereafter, was prompted by the search toreduce public spending and the growth of statutory services. Contracting with third-sector organizations offered the hope of lower costs. Also, con-servative governments came to power in the UK, the US, New Zealand,Australia and elsewhere that were decidedly more skeptical of public services

and more supportive of market approaches. Consequently, public agencies(usually local government in countries such as the UK, Australia and NewZealand) were pushed to transform their role into one of a purchaser andenabler rather than a direct provider of services (Gutch 1992; Considine 2003;Lyons forthcoming). This shift to the enabler role fit with the emergence ofthe “new public management” movement (Hood 1991; Rhodes 1996) thatstressed “steering rather than rowing” for public agencies (Osborne and Gaebler1993). To varying extents, this shift in policy is evident in many other countries such as the UK (Kirkpatrick et al. 2001; Gutch 1992), the US (Smithand Lipsky 1993; Gronbjerg 1993), Finland (Simonen and Kovaainen 1998),and Australia and New Zealand (Considine 2003; Lyons forthcoming). Inthese countries, this change to formal contracting has also entailed greaterreceptivity to government contracting with for-profit service agencies espe-cially for long-term care and childcare.

In the UK, the government has pledged to increase the involvement ofthird-sector organizations in the provision of public services. Indeed, in 2006,the UK government formally asserted its intention to increase the involve-ment of the third sector in the delivery of public services (Cabinet Office 2006).This new target reflects a number of policy objectives:

• to introduce contestability and choice into public services by increasingthe quality and range of organizations that can be commissioned to provide services;

• to ensure that the statutory sector is able to access the expertise of thethird sector in delivering services to communities, particularly harder-to-reach communities and challenging client groups;

• to increase the capacity of public services as a whole to respond to grow-ing demand for services and increased client choice; and

• to increase social capital within communities and facilitate the engage-ment of service users (Office of Government Commerce 2006).

In response, a new emphasis on commissioning as the central function oflocal government has developed, especially in England2 as described in theLocal Government Act of 2006. Commissioning, in this context, is the wordused to describe the activities and processes used by governments to decidehow best to use the total resources available to a population in order to improveoutcomes. As applied to contracting, commissioning entails working with andthrough a range of different organizations, including third-sector organiza-tions, to deliver better services for the public. At its core is an understandingthat commissioners are purchasing services from public service teams withinthe local authority as well as from arm’s-length provider organizations suchas schools and hospitals, third-sector organizations, and for-profit providerbusinesses. Government policy-makers also hope that commissioning is aneffective strategy to ensure that all providers of public services are contestible,

even when formal competitive tendering has not been used (Audit Commis-sion 2008). As articulated in the government’s White Paper, local authoritiesshould play a coordinating role, facilitating cross-agency collaboration andcooperation to produce innovative, joined-up services involving public- andthird-sector organizations, especially for vulnerable people. Many policy-makers and practitioners in the UK are seeing the future role of local govern-ment as commissioning from a range of service providers – only includingthose provided from within the public sector if they are evidentially betterand more efficient than potential competition.

In practice, though, the shift to contracting in many countries such as theUK and the US has not led to widespread privatization of public services.Many of the services provided by third-sector organizations are relativelynew. Thus, contracting has been used to respond to innovation and createentirely new services, frequently in response to new social movements suchas the women’s movement. For instance, in the US, the UK and Canada,domestic violence programs are provided by voluntary agencies under govern-ment contract. Many other examples exist, including immigrant assistanceand AIDS services. In the UK, national program such as Surestart, theChildren’s Fund and Connexions provided new opportunities for third-sector providers to offer new services, and in some areas the third sectorincreased its role as service provider quite considerably – although a frequentcomplaint of third-sector organizations is that they are used when fundingstreams are short-term and uncertain because it is easier to discontinue anexternal contract than to lay off government employees.

The political attractiveness of contracting and the use of contracting as a quick way to respond to new needs is evident in the growth of contract-ing among governments and large multi-national NGOs such as Save theChildren, CARE and Oxfam. These third-sector organizations receive hun-dreds of millions of dollars in contracts to provide humanitarian relief anddevelopment programs. In addition, many countries in the developing world have contracts with locally based NGOs for a variety of importantservices. The extent of contracting in the developing world has exploded in the last twenty years owing to the changing character of foreign aid; recurring crises, wars and natural disasters; and political liberalizationallowing the growth of the NGOs.

In the UK, these aspirations for greater engagement of the third sector in providing public services through contract has been generally welcomedby much of the third sector (Etherington 2006). Nonetheless, little evidenceexists to suggest that significant redistribution of work from the public sectorto the third sector has occurred despite specific targets for third-sector pro-vision. For example, a Public Accounts Committee report published in 2006,Working with the Voluntary Sector, found that the government had failed tomeet key targets to increase the third sector’s involvement in public services(Public Accounts Committee 2006). The committee recommended that new

targets should be established beyond 2006 to provide a real incentive for depart-ments to increase their involvement in the sector. Indeed, the creation of children’s and adult services authorities has resulted in work previously contracted out to the third sector transferring back into local authorities(including SureStart, the Children’s Fund and Connexions services). And itshould be noted that the UK government’s targets to increase the propor-tion of services provided by the third sector are in conflict with governmentpolicy to develop local-authority commissioning capacity which is focusedon choosing the best provider (in terms of willingness to change, efficiencyand impact on outcomes), regardless of sector. Commissioners in Englandare being trained to focus on outcomes and to be sector-neutral. Arguably,the threat of privatization is acting as a spur for innovation and serviceimprovement within the public sector as suggested by recent changes andreforms in adult and children’s services in the UK.

One additional trend is evident in the relationship between governmentand the third sector. In the last ten years, many countries have implementedpolicies to individualize and personalize services. In some countries, the resultis new consumer subsidies as a strategy to provide needed public services.Many countries offer childcare vouchers for eligible parents that the familycan use to pay for childcare from any eligible provider (Adams, Rohacekand Snyder 2008; Simonen and Kovalainen 1998). And many countries increasingly employ housing vouchers to help the poor and disabled affordadequate housing (Priemus 2000).

Further, in the UK, the US and elsewhere, policy-makers, service-providers and consumers have individual budgets to personalize services forthe aged and for people with disabilities through greater individual controlover service decisions. Typically, the service recipient makes the decision on service providers with the help of a public service broker. Some recentstudies show enormous benefits for individuals and considerable savings tothe public purse.3

While both vouchers and individual budgets for social and child care andhousing are not typically earmarked for third-sector organizations, they havehad important direct and indirect effects on government–third-sector rela-tionships. Individual budgets and vouchers call for new funding arrangements.Commissioners need to know the unit cost and overall value of services, producing an incentive for government administrators to create new contractforms including more spot purchasing for specific clients or circumstancesand less reliance on potentially wasteful block contracts that provide fund-ing for a group of clients and services. Skilled commissioners will need tounderstand how each provider is financed and the critical balance betweenfixed and variable costs. In general, individual budgets and vouchers encour-age greater competition among service agencies and greater uncertainty amongproviders on receiving payments from government. However, individualbudgets and vouchers tend to shine the spotlight on the quality and impact

of services and encourage service providers to market their services directlyto potential clients and the public in ways that could potentially lead to greaterresponsiveness to clients in the future.

Contracting in practice

The expansion of government contracting with third-sector organizations hascreated new challenges and opportunities for these organizations and, morebroadly, the governance of public services. These challenges can be groupedinto three broad categories: performance contracting and management; self-regulation, inspection, quality assurance and accreditation; and competition.Each of these factors has a substantial and enduring effect on the governanceof services provided by third-sector organizations including the respective roles and responsibilities of public managers and the boards and staff of third-sector organizations. As will be discussed, these developments and challengessuch as performance contracting can undermine the potential advantages of third-sector organizations that attracted public managers to engage theseorganizations in public service delivery through contracts. Also, these chal-lenges and the effect on the third sector will also be evaluated at the end of each section with respect to the standards for good governance for thepublic services developed in 2004 by an independent commission jointly sponsored by the Office for Public Management (OPM) and the CharteredInstitute of Public Finance and Accountancy (CIPFA).

Performance contracting and management

To varying degrees, the growth of government contracting with third-sectororganizations has produced increasingly intensive efforts to hold these organizations to account. This effort began with the initial build-up of government contracting. Service agencies were expected to be accountablefor their expenditure of public contract funds. However, this accountabilitytended to be in a line-item sense: Were the public funds spent according tothe stipulations of the contract? This “process” or “regulatory” accountabilitytypically involved reporting the number of clients served or meals deliveredas well as a budgetary breakdown on the allocation of various expenditures(Behn and Kant 1999). Government also tried to increase the likelihood that the contract agency delivered a quality program by requiring staff andprogram standards even before any money was distributed. This objectiveis often achieved by a pre-qualifiying questionnaire in which evidence of accreditation, governance and financial solvency are required. In this initialperiod, accountability expectations tended to be relatively modest and/or informal.

Over time, many policy-makers became disenchanted with this processaccountability, arguing that it provided little evidence of the effectiveness with

which programs were achieving their desired outcomes. Further, govern-ment administrators worried about the potential mismatch between public-and third-sector priorities. As a result, public funders began to require that service agencies measure the impact they are having on outcomes for service users and taxpayers. Toward this end, government funding agenciesinstituted, to varying degrees depending upon the country and jurisdiction,performance-based contracting that strives to hold nonprofit (and for-profit) agencies accountable for reaching specific performance targets (Behnand Kant 1999; Forsythe 2001). Indeed, many governments, including theUS, the UK, Australia and New Zealand, will not reimburse third-sector organizations unless they meet performance targets, although the extent andrigor of the performance contracting varies depending upon the service category and prevailing public policies (Smith forthcoming; Lyons forthcoming;Phillips forthcoming).

Outcomes-based contracting is being developed across the world. Ratherthan specifying activities and processes, commissioners are now working out how best to specify outcomes. Thus, third-sector service providers areincreasingly expected to design the optimal service for the money availableand demonstrate success through evidence of the impact on outcomes.Toward this end, in the UK, many policy-makers and administrators have embraced, the framework of Results Based Accountability by MarcFriedman (2005) and similar approaches that focus commissioners and ser-vice providers on outcomes. Friedman initially developed his ideas throughwork with federal and state governments in the US. His work is already beginning to be reflected in the nature of contracts in the UK and other countries, including New Zealand, which will over time change the way inwhich third-sector organizations measure and manage performance internallyand the focus of attention of their boards.

Better performance by service agencies was of course a central goal of manypolicy-makers interested in expanding government contracting with nonprofitorganizations; thus, the widespread application of performance managementto third-sector organizations receiving public funds is a logical development.However, this heightened interest in performance has a number of implica-tions for the governance of third-sector organizations, including the role of voluntarism within these agencies. This shift to outcome evaluation ofteninvolves a revolution in thinking for agencies and their overall management.Agencies need new investments in management information systems and monitoring in order to track and compile important programmatic and financial data, track outcomes (to the extent feasible), and better understandtheir cost structure. The effect is to “professionalize” the administrative infrastructure of third-sector organizations that can then create problems ofmission and focus for an organization since a more substantial infrastruc-ture can require new resources that may be at variance with the previousprogrammatic focus of the organization. However, professionalization can

also be beneficial for the organization because it may allow the organiza-tion to manage its finances and programs more effectively. Overall, a moreprofessionalized infrastructure may be required in order to sustain the organization and have a positive impact.

Moreover, in the UK, government is striving to standardize the assess-ment and referral forms used by all service providers in many services, including social housing, children’s and adult services, in order to facilitatepersonalization and the collection of evidence of impact on outcomes fromall the different service providers in these service categories. For example,children with additional needs will not be able to access the range of serviceswithout a Common Assessment Form (CAF) and a lead professional. Third-sector organizations will need to complete these forms and act as lead professionals, which will further standardize their work since compliance with these regulations will be required.

These adaptations can be an especially difficult challenge for smaller community-based organizations, given their relative undercapitalization andspecialization. Many third-sector organizations started through commu-nity initiative often lack, at least initially, a highly trained professional or administrative staff. Some of these agencies originally emerge from an unincorporated group of people concerned about a social problem. A clearseparation between the board and staff is absent, and many of these agenciesdo not have full-time executive directors. Government accountability measures,including performance-based contracts, often require these agencies to adoptnew administrative procedures, add professionals, institute new financial management practices and, in some cases, modify existing physical structures.Consequently, the voluntarism that is the hallmark of many community agencies in their start-up phase is difficult to sustain in the current environ-ment with the pressure to be more accountable to government and donors.Given this current emphasis on performance, third-sector organizations thatcontinue to rely upon volunteers in direct service roles tend to be programsoffering low-intensity services such as emergency assistance, crisis hot-lines,and tutoring.

Importantly, commissioners and third-sector organizations in the UKand the US have encouraged coalitions of third-sector organizations tocome together for the purposes of contracting. This effort is challenging to the independence of third-sector organizations and can undermine theirgovernance even when great care has been taken to agree clear accountabilityarrangements in writing.

This shift to professionalization and performance contracting can beespecially consequential for the relationship between third-sector organiza-tions and their clients. Many third-sector organizations emerge out of a desireof a “community” of people to address a problem or social need such ashomelessness, arts education, or workforce development. These individualscreate a service agency that regards its mission as logically being responsive

to their community of interest (Smith and Lipsky 1993). Government, bycontrast, tends to approach services and clients from the norm of equity, con-sistent with the need of government officials to treat groups and individualsfairly. Equity can be interpreted in a variety of ways, but in social and healthservices it usually means defining need in order to allocate resources by criteria deemed to be fair – e.g. income, geographic location, and severity of illness or need. Because of their emphasis on responsiveness, third-sectoragencies may clash with government, especially on policy matters relatingto services, clients and staff. This clash can be especially pronounced undera performance-contracting regime which can leave third-sector agencies littlediscretion on the performance targets to be met and may require the agencyto shift its programmatic focus toward short-term goals and client groupsat variance with their original community of interest (Smith and Lipsky 1993).Likewise third-sector organizations established to deliver a particular pro-gram of care (such as a “nine-step program” or particular form of therapy)will be challenged by outcome-focused contracts. In the past, they have woncontracts from commissioners persuaded of the efficacy of the approach. Inthe future, they will need to demonstrate impact on outcomes, and in somecases they will have to consider a range of therapies in order to hit targets.If they fail to demonstrate target outcomes, they could be marginalized by commissioners who perceive that the third-sector providers are more concerned with process than with outcomes.

Another difficult program-related dilemma raised by performance con-tracting and contracting within competitive environments is that agencies may face incentives to “cream” the clients who have the greatest possibilityof successful outcomes. For instance, performance-based job-training contractsmay encourage agencies to work with clients who can succeed, especially sincean agency will not be reimbursed unless clients are successful. In a com-petitive contract culture, agencies want a positive reputation in order to obtainnew contracts or retain existing contracts. This incentive can promote the creaming of clients, even if it is counter to the overall objectives of thecontract (see Considine 2003; Behn and Kant 1999).

The government–third-sector organizational relationship may be differentif the contract is regarded as a one-time event. In the US, some federal con-tracts to third-sector agencies are awards for program innovation with noexpectation of a continuing contract. However, even in these circumstances,the federal government increasingly requires extensive documentation of pro-gram services and finances by the agency. Also, the competition for contractfunds means that, even if a contract is a one-time event, the agency may hopeto replace a contract with another contract. For example, substance-abuseagencies have been able to obtain federal grants sometimes through theDepartment of Justice for innovative treatment programs for parolees.However, these agencies often hope that state government will assume theresponsibility for funding these services or another federal contract will be

awarded to take the place of the expiring contract (Sosin, Smith, Hilton andJordan forthcoming).

Third-sector agencies may try to overcome some of the limitations imposedby government contracts by using private funds to cross-subsidize the costof service. Or workers may essentially volunteer their own time to help par-ticular clients in ways that go beyond the expectations of the governmentcontracts. Yet the more recent shift in government funding toward perform-ance contracting and standardization makes it more difficult for agencies togo beyond the expectations of the contract. Low vendor rates and greatercompetition also push agencies to be more bottom-line-oriented and reducethe “slack” in agency operations. As a result, agency staff are likely to interact with clients differently and focus even more directly on the specificexpectations of the government funder.

Importantly, another key trend related to the heightened interest in performance is the growing interest and support among policy-makers andadvocates for greater personalization of services. In the US, this trend isreflected in the creation of individual accounts for disabled children and adultsthat allow them or their guardians to choose their providers and thus directtheir public funding.4 Increased funding for vouchers in housing and child-care, rather than direct contracts to providers, also exemplifies this trend. Inthe UK, the Department of Children and Families (DCFS) is experiment-ing with lead practitioners or “brokers” with budget-holding authority whoexercise choice in terms of services on behalf of children and families (DCFS2006; Gillanders 2009). As in the US, the UK government is also establish-ing an individual budget for the disabled (Prabhakar, Thom, Hurstfield andParashar 2008). However, this movement toward individual choice places evengreater pressure on government to ensure the quality of local providers. Localcommissioners and public managers will need to know that providers meetlocal standards of service and management, including safety and governance.Over time, the shift to vouchers and individual budgets will undermine thetraditional contract relationship between government and third-sector pro-viders. It has the potential to improve outcomes by creating more informedmanagers, purchasers and citizens, but it could also create greater financialand programmatic uncertainty among providers since they will now have to compete for funding with individuals and/or their guardians.

Interestingly enough, the advent of performance contracting and the morerecent push for greater choice in services may encourage more commu-nity connections by many third-sector agencies. Many agencies founded inthe last thirty years were started by individuals who were very passionateabout addressing a specific problem, such as economic development, at-riskyouth, substance abuse, housing, or mental health. The boards tended to be small, and government contracts tended to encourage passivity in boardgovernance since many board members were not familiar with the intricaciesof government contracts and joined the board for their commitment and

support for the agency’s clients and programs rather than for their fundingexpertise and connections. Further, since funding was initially adequate tosupport operations, agency boards had little incentive to seek broad-basedcommunity support. But the new pressure to be efficient and more account-able pushes third-sector community agencies to restructure their boards andtheir community relations in order to improve their competitive position for public contracts and to raise private funding. Agencies also want to havebroad community support in order to enhance their profile and perceivedlegitimacy, particularly given the greater importance of client choice.

In conclusion, set against the six standards described in the Good Govern-ance Standard for Public Services (Independent Commission 2004), the effectof performance contracting on governance is summarized in Table 16.1.

Self-regulation, inspection, accreditation and quality frameworks

The dilemmas and challenges of performance- or outcome-based contract-ing between government and third-sector organizations has encouragedgovernments and third-sector organizations across the world to develop newapproaches to ensuring service quality and good performance, including greateremphasis on technical assistance and support, self-regulation, inspection andaccreditation.

Broadly speaking, technical assistance is often delivered through a varietyof different capacity-building initiatives. In the UK and the US, four different approaches to improving the capacity and professionalism of thethird sector are evident. First, governments have specifically encouraged thebroader application of accreditation policies. In the UK context, accredita-tion means that organizations or individuals are accredited to deliver a particular service, and that accreditation indicates that they have “achieveda specified, minimum standard, laid out by the accrediting body” (Departmentof Health 2006a: 34). Overall, an accreditation process is a one-off measureof current competence; if an organization is successfully accredited, then it may be awarded a “kitemark”. For instance, the acquA accreditation process was developed by third-sector organizations in Herefordshire, UK,and is administered by the Health and Social Care Alliance. Commissionersfrom the NHS, and local government have agreed not to contract with any organization not accredited through acquA.

Second, in the UK, a strong tradition of regulation and inspection has developed. External bodies set standards and carry out inspections with theaim of assessing whether organizations are meeting the minimum standardsrequired. Regulatory bodies and inspectorates for health and social care in the UK include the Health Care Commission and the Commission forSocial Care Inspection (CSCI). In the US, these regulatory bodies such asthe Commission on the Accreditation of Rehabilitation Facilities (CARF)and the Joint Commission that accredits healthcare organizations are also

lifica

s’ r

increasingly important as a way for government to encourage service qualityand better performance by private providers.

Third, in the UK and the US, growing interest exists in developing qualityframeworks specifically tailored for the third sector. A quality framework isa structure for examining the quality of an organization across a number ofareas. For example, the UK’s Charity Commission’s Hallmarks describe sixoverarching principles of an effective charity that together creates a frame-work for effectiveness to guide charities in their work. Quality frameworksare often used to promote organizational learning and development. Goodexamples in the UK include Investors in People, Quality Counts and the GoodGovernance standard for public services which underpins the governanceframeworks for local government, Children’s Trusts, and wider public servicesin England and Wales (Audit Commission 2008). In the US, the MarylandAssociation of Nonprofits has developed the Standards of Excellence whichdetail good governance standards for nonprofits (Maryland Association ofNonprofits 2009). And the Panel on the Nonprofit Sector (2007) sponsoredby the national organization, the Independent Sector, issued a report entitledPrinciples of Good Governance and Ethical Practice as a guide for self-regulation by nonprofits and foundations.

In the UK, peer review is also increasingly regarded as an important strategy to enhance the performance of third-sector organizations. These pro-cesses involve external experts in the relevant field assessing an organization’sperformance or providing challenge and support to an organization. Forinstance, the Improvement and Development Agency for local government(I&DeA) supports peer review to promote better performance in local government, including commissioning with the third sector (I&DeA 2008).Peer review may involve an individual or team of peer reviewers from within the third sector carrying out an assessment of how well an organiza-tion meets the requirements of a quality framework. Another variation thatis gaining greater credence is peer referencing whereby individuals or teamsof reviewers assess an organization and benchmark it against other com-parable organizations in its sector or field. This type of formalized peer reviewwithin local government or the third sector tends to be less advanced in mostother countries, including the US and Canada.

Peer review and the broad interest in quality standards and frame-works such as the Standards of Excellence are part of a widespread moveto enhance the capacity of the third sector for self-regulation. For instance,the Charity Commission in the UK (2006) explicitly states a commitment to encouraging self-regulation and a reduced regulatory burden where pos-sible in their Partnership strategy. The Commission states that work currentlyongoing to help large umbrella charities review their quality standards tools to reflect the Commission’s Hallmarks of an Effective Charity couldlead the Commission to withdraw from conducting Review Visits to thesecharities.

As another indication of the commitment of the UK government toenhancing performance management and regulation in the third sector, theTreasury conducted a cross-cutting examination of the role of the third sector in service delivery as part of the 2002 UK government’s Spending Review(HM Treasury 2002). The substantial contribution that the third sectoralready made to public service delivery underpinned the review, which thenproduced new government policies regarding the third sector and commis-sioning. First, new initiatives such as FutureBuilders and CapacityBuildersare designed to increase the capacity and professionalism of the third sectorin order to help it meet the demand for more public service provision.5 Second,government is trying to reduce the barriers in current procurement, com-missioning and contracting regimes faced by third-sector organizations inaccessing contracts or funds. (Similar initiatives have occurred in the US.)Third, government and third-sector organizations are at least rhetorically committed to the goal that third-sector organizations engaged in public service delivery should be well governed and managed, properly regulated,and meet at least the minimum programmatic, managerial and ethical standards (NCVO 2008).

The tension between government accountability expectations and self-regulation and control by third-sector organizations that is evident in manycountries has been especially evident in the UK in recent years. For example,the 2002 UK Treasury review commented:

in contracting with the voluntary community sector to deliver services,Government must ensure that regulation is proportionate and theindependence of the sector is recognized; the greater the regulation,the greater the risk that the best features of the sector are smothered.

(HM Treasury 2002: 17)

In more recent policy statements, such as the Report of the Third SectorCommissioning Task Force (Department of Health 2006), the Govern-ment stressed the need for the third sector to demonstrate their “fitness forpurpose”. However, the report also warns that “disproportionate andinconsistent demands of multiple regulators” presents a major barrier toincreasing the involvement of the third sector in service delivery. The reportalso suggests that a simplification of the present regulation system is needed,commenting that:

Regulation, licensing and accreditation of health and social care providers needs to be equitable, proportionate, risk-based andstreamlined, so that it optimises the overall inspection and moni-toring burden proportionate to the balance of users’, providers’, commissioners’ and regulators’ needs.

(Department of Health 2006b: 17)

The report also highlights the merits of peer-review initiatives such as acquAto enable the third sector to meet the demands of contracting with public-sector organizations.6

To streamline regulation further, the UK government’s Better RegulationTask Force (BRTF) (2005) issued a report entitled “Better Regulation for CivilSociety”, which recommended that the UK funding and regulatory govern-ment departments undertake a systematic measurement of the administrativeburdens associated with third-sector contracts. The principle of propor-tionality in regulatory systems is particularly pertinent for the third sector,meaning that a regulatory system does not treat all organizations the samebut is proportional to the organization’s size and capability to respond toregulatory requirements.

Another vitally important influence on third-sector organizations in theUK is the Compact: a formal agreement negotiated between the governmentand leading third-sector organizations in 1998. It is intended to apply to theinteractions between government at all levels including local government andthe NHS and the third sector. After a slow start, most local commissionersare implementing the basic principles of the Compact whenever they are contracting with third-sector organizations. Indeed, the Commissioner forthe Compact is considering plans to develop a potential accreditation for good Compact practice (Compact 2008).

The UK Compact has generated broad attention throughout the worldfrom governments and third-sector organizations. Some countries, such asAustralia, have experimented with local-level compacts (Casey and Dalton2006). However, no country has yet developed as structured an agreementand relationship as the UK. The reasons for the divergence between the UKand the rest of world relate to its strong national government, extensive voluntary sector, and relatively weak local government. Further, many lead-ing voluntary organizations in the UK are highly dependent upon state support – unlike in the US where the extent of government support varieswidely. Other countries such as Germany and the Netherlands already havecorporatist-like structures in place that permit and encourage bargainingbetween government and the voluntary sector, mitigating somewhat thepressure to create new institutions and structures like the Compact.

In sum, contracting in the UK, the US and other countries appears to beentering a new uncertain phase: greater interest in accountability and perform-ance but also more support for decentralization, discretion and individualchoice. In this sense, the tensions evident in contracting with third-sector organizations embody some of the inherent contradictions of the New PublicManagement more generally (Rhodes 1996) which promotes markets and better performance and more community and citizen responsiveness. The push for accountability is creating stricter standardization of expectationsby government of third-sector organizations. Board members and staff ofthird-sector organizations may thus feel very disaffected because of the lack

of independence and the burdens of compliance. One could argue that theendgame is the transfer of better governance practices to the board of third-sector organizations and the consequent reduction of external regulation and inspection. However, this shift seems unlikely. Meanwhile third-sectorprovider organizations in many countries, including the UK, the US, Australiaand Canada, face the need to be increasingly compliant and accountable ineverything they do.

In conclusion, set against the six standards described in the Good Govern-ance standard for public services (Independent Commission 2004), the effectof increasing external regulation and inspection on governance is similar in some respects with the impact of performance contracting. These effectsare summarized in Table 16.2.

Increased competition

The growth of performance contracts and the more general interest inenhanced accountability promotes greater competition for public contractsand private donations, although obstacles to effective competition and broadercounter-trends at the street-level promoting collaboration also make the imple-mentation of effective competition practices difficult. In the initial build-upof government contracting with third-sector organizations, most agencies didnot really compete with other agencies for contracts. In the US and in othercountries, most contracts were cost-reimbursement contracts that essentiallypaid agencies for their costs based upon the contract terms and budget.Reimbursement was not linked to outcomes, and most agencies recoveredtheir costs (at least as specified in the contract). Little incentive existed foragencies to compete with other agencies for contracts since contracts wereunlikely to be moved from one agency to another unless egregious problemsexisted. Moreover, outcome data were lacking that would provide a basisfor changing contract agencies. The advent of performance contracting givesgovernment much more information and justification for changing contractagencies if performance targets are not being met. At least some third-sector organizations also have an incentive to compete with their fellow agencies since they could potentially grow through additional contracts.

The increase in competition for public and private grants (and clients) isalso related to other important trends affecting third-sector agencies. Publiccommissioners are seeking to provide a more level playing field for serviceproviders from different sectors, creating more competition for third-sectororganizations from for-profit and public-sector teams, although the extentof this competition varies greatly depending upon the jurisdiction.

Thus, for-profit firms now compete for contracts in service categories previously dominated by third-sector providers, including childcare, homecare, and community programs for the mentally ill and developmentally disabled. For-profits may possess some advantages vis-à-vis nonprofits in the

competition for contracts and government funding. First, for-profit chainshave access to capital and a sufficient size that allows substantial economiesof scale, allowing it to operate at least some programs more efficiently. Second,many third-sector organizations are mission-based and small and unwillingto serve certain type of clients or in specific regions, reducing the oppor-tunities for them to cross-subsidize their operations through growth or adiversified client mix. Many community-based organizations may also be veryambivalent about expansion (or lack the capacity for growth). For-profitstypically do not have these types of mission constraints and, consequently,may be more willing to serve a diverse mix of clients, including controver-sial clients. Third, for-profits tend to be newer entrants to the provision ofsome types of services such as home care or community programs for thementally ill. As a new entrant, they may be able to obtain substantially higherrates for their services than a thirty-year-old community-based organizationsince rates for specific agencies tend to be quite dependent upon the date offounding (and the negotiating skill of the chief executive). Once established,rates tend to grow incrementally, and during periods of fiscal crisis or austerity actually decline.

The variation in the mix of for-profits and third-sector organizations in particular service categories in different localities reflects several factors.Some jurisdictions discourage or restrict the entry of for-profits. In some areas, the third-sector provider community is so large and entrenched thatfor-profits are themselves discouraged from entering specific service markets.For-profits may avoid financially risky services such as serving the homelessmentally ill. Further, rates for some services may be so low that for-profitsmay find it difficult to make a profit; hence nonprofits that can cross-subsidize their services with private donations or earned income may havea competitive advantage.

A second important trend facilitating greater competition is the push forgreater client choice in the selection of service agencies, especially in com-munity care for the aged and disabled. To the extent that clients havegreater choice of services, agencies will need to adjust by marketing their services to citizens and competing with other agencies. The movementtoward client choice varies significantly between countries and even withincountries. Nonetheless, this trend is especially noticeable in services for theelderly, MR/DD clients, and the chronic mentally ill that offer them moreindependence and control over their own lives. While client choice in serviceproviders remains limited in some service categories and jurisdictions, it isslowly building momentum and influence, so it is likely that greater choicewill be available in the coming years.

The third factor promoting competition is the sheer number of third-sector organizations in many communities. Many are relatively small andyoung. So they are now at a point in their organizational life-cycle wheresome of the initial grants and contracts are expiring or maturing. So these

organizations need to create a sustainable plan for the future that includesnew sources of revenue. This planning often involves competing for publiccontracts and grants.

Finally, the increasingly intensive accountability measures including per-formance contracts can intensify the competitive pressure on nonprofits.Specifically, performance contracts raise the specter that an agency could lose its contracts if it did not meet specific performance targets. Relatedly,governments in the UK, the US (Smith forthcoming) and elsewhere haverestructured many contracts from a cost-reimbursement to a fee-for-service,allowing government to exert greater fiscal and programmatic control overcontract agencies (at least theoretically); in effect, contracts and funding fromgovernment (and private funders) is increasingly restricted, creating greateruncertainty and prompting agencies to be more aggressive in seeking newfunding sources. This trend can be counter to other strands of governmentpolicy which stress the need for long-term partnerships in which contract relationships are open, transparent and characterized by continuous improve-ment on both sides.

The greater competition for public contracts (and private donations, depend-ing upon the country) encourages third-sector organizations to tap earnedincome and fees more extensively as a source of revenues. In the context ofsocial and health services, fees and earned income can involve a variety of diverse revenue options: the payment of fees by individuals, public andprivate insurance companies, and corporations; selling services such as technical assistance or cookbooks; earning money from client-run businesses,such as a restaurant staffed by the disadvantaged; real-estate development(including parking revenue); and selling food at a local festival.

Despite the broad attention to earned income and fees by third-sector organizations (Alter 2007), many third-sector organizations are not well positioned to raise substantial earned income through the sale of services(Foster and Bradach 2005). Many community organizations are, as noted,undercapitalized, limiting their ability to launch new projects or initiatives.They also tend to be very value-driven, with scant interest in expanding their earned income. Also, many agencies do not have any service that theycan profitably sell on the “market”.

The growing emphasis on earned income and competition raises a broaderconcern that many community-based organizations may feel pressure to bemore competitive and emphasize services that generate adequate public andprivate revenues, thus shifting away from their commitment to support-ing valued community services lacking adequate funding (Eikenberry andKluver 2004; Alexander, Nank and Stivers 1999). In the process, nonprofitsmay change their governance structure through staffing and board changesthat diminish their commitment to longstanding values and their historicalmission. However, some third-sector organizations may deepen their con-nections to their community in response to competition since agencies have

an incentive to broaden their community and political support to aid themin successfully winning contracts, obtaining private grants and donations,and developing earned income sources (which often need private grants fortheir initial capital).

Overall, the trend is certainly toward greater competition among con-tract agencies; however, other developments are encouraging nonprofits to consider greater collaboration with other public and private organizations.The economic crisis is forcing increased consideration of collaborationamong service agencies to reduce costs – including, in more extreme cases,full-scale merger. Faced with fewer financially stable agencies, governmentofficials may feel pressure to work more collaboratively with third-sector service providers. And some localities lack a sufficient number of third-sector providers to have effective market competition for government contracts. Also, contracts themselves have become more complex and mayinclude requirements for collaboration among service agencies. For example,a program for at-risk youth may be delivered by a third-sector organizationin collaboration with the local school district.

In conclusion, set against the six standards described in the Good Govern-ance standard for public services (Independent Commission 2004), the effectof competition on governance is summarized in Table 16.3.

The politics of contracting

Prior to extensive contracting, agencies tended to focus on their own pri-vate set of concerns that were regarded as quite separate from the publicsector, and government regulation was quite minimal. Consequently, agencies possessed little incentive to actively lobby government (especially since government was unlikely to respond with any substantial changes in policy).However, contracting profoundly changed the incentives faced by third-sector organizations regarding political activity and their interest in influencinggovernment policy. The availability of contracts meant that agencies couldobtain funds for new and existing programs. And agencies receiving govern-ment contracts were directly affected by government policy regarding con-tract rates, regulations and programmatic standards, referral policies and the overall budget climate. As a result, third-sector organizations are nowin a different political position than prior to the contract era, with much greater incentive to engage in the political arena and to strive to influencepublic policy.

Nonetheless, third-sector organizations can face significant obstacles to political activity and advocacy. As noted, many agencies are quite small andlack the staff resources actively to participate in the political process. Theboard members of many third-sector organizations tend to be attracted to board service owing to their commitment to the agency’s mission and services such as child welfare or homelessness; consequently, most board

members possess little experience with government policy. In many countries,the board and staff of the organization may also be concerned that polit-ical activity might jeopardize their relationship with government and, in theUS, even their tax-exempt status.7

In response to these sometimes contradictory pressures, third-sector organizations have an incentive to construct their political strategies with arecognition of the potential impact on their relationships with governmentand the resource constraints facing these agencies. First, third-sector organ-izations in many different countries, including Canada, the UK, Australiaand the US, have banded together to create coalitions and associations ofproviders to lobby government and advocate key agency priorities. Thesecoalitions can be formally incorporated tax-exempt organizations with paidstaff or more loosely organized associations with no formal legal status. Somecoalitions are sizable enough to have paid staff while other coalitions haveno paid staff by the coalition, although they may depend upon the in-kindcontributions from the member agencies.

In the UK, two of the most widely known associations representing thethird sector are the National Council of Voluntary Organisations (NCVO)and the Association of Chief Executives of Voluntary Organisations (ACEVO).In addition, there is a range of sector-specific bodies such as the NationalHousing Federation and an emerging group of infrastructure bodies repre-senting, for example, hospices, foundation hospitals and academy schools.In the US, national-level associations include the Alliance for Children andFamilies and the National Council of Nonprofits. Many state-level coalitionsand associations also exist, including the Minnesota Council on Nonprofitsas well as subsector organizations such as the Massachusetts Association ofHuman Service Providers.

A second response of third-sector agencies to the politics of contractingis to hire staff or appoint board members with experience with governmentcontracts. In many jurisdictions, third-sector agencies now have executive direc-tors who previously held important government positions. Board memberscan now include prominent local leaders with credibility and legitimacy to influence government policy. Of course, individual agencies will vary inthe extent to which they have staff with previous government experience; in general, the larger agencies with the more extensive resources are in a better position to compete for these highly desirable staff and executives.

Hiring staff with government experience is a very public acknowledgmentthat government contracting can be a long-term relationship for manythird-sector organizations. Today, third-sector organizations provide anincredibly diverse array of services, often under very difficult circumstances.Many government contracts are for ongoing services such as child welfare,prisoner re-entry programs, community programs for the development-ally disabled and mentally ill, home care for the elderly, and workforce development. To be sure, the certainty of funding is often very unclear and

unpredictable. Many agencies, especially larger agencies, may experiencedeclines in some contracts and increases in other contracts in any particularyear (Sosin, Smith, Hilton, and Jordan forthcoming; Allard 2008). However,agencies tend to have long-term relationships with government and areexpected to be able to weather the cyclical nature of government funding.As a result, third-sector organizations need to invest in their staff with the goal of building productive long-term relationships with government in order to be effective and sustainable; staff with government experiencecan be of great assistance.

Thinking about the relationship with government as a long-term relationshipis reflected in other organizational adaptations to contracting. Governmentcontract administrators place great priority on several key managementcomponents: sound financial management, including regular audits; atten-tion to performance management, including the use of current program evaluation models; up-to-date tracking of clients and services; and a goodreputation in the local community. In order to meet government expecta-tions on these important priorities, third-sector agencies with contracts (orinterested in obtaining contracts) need to invest in expertise in financial management, program evaluation and information technology. Over time,then, government contracting tends to change the internal agency dynamicsand management style profoundly.

The movement of professionals back and forth between government andthird-sector contract agencies tends to encourage a consensus on program-matic standards in particular service categories. Over time, this consensuscan become a characteristic of the “contracting regime” whereupon governmentand third-sector agencies develop a common set of assumptions to guide theirrelationships. The regime concept suggests that two parties are mutually dependent upon each other, so that each party cannot easily leave the rela-tionship. However, an equally important aspect of regimes is that one partyis typically much more powerful than the other. In the case of contractingregimes, government tends to be the more powerful partner and is in a position to dictate programmatic and financial expectations even in the face of opposition from their contract agencies (Smith and Lipsky 1993; also Considine 2003; Considine and Lewis 2003). For example, a public child welfare department and a set of third-sector child welfare agencies may develop specific norms about acceptable practice, referral policies andreimbursement rates. These norms then guide the behavior and strategic management of the government department and the providers. In this relationship, government is able to drive the evolution of these norms, giventheir resources and political influence and the relative absence of alternativefunding sources for their programs.

Contracting regimes tend to operate based upon a certain level of trustamong the two parties, despite this power imbalance. Contracts are oftenrelational and long-term even in instances where competitive bidding is

required (Smith and Smyth 1996). This “relational contracting” does not preclude differences of opinion or outright conflict but it does underscorethe stability of many contracting arrangements and the importance of cooperation among the two parties (Deakin and Michie 1997; Ring and Vande Ven 1992; Van Slyke 2007). But this cooperation occurs within a frame-work established by the more powerful partner. Third-sector organizationsmay cooperate with government on a contract for community care for thechronically mentally ill; but the standards of care, financial regulations and outcome measures are still set largely by government. Put another way,a third-sector organization, by its decision to contract, indicates a willing-ness to cooperate with government on the implementation of a government program. Consequently, relational contracting does not capture the natureof typical contracting relationships since “relational” implies equity in decision-making and power.

In the process, third-sector organizations tend to adopt similar internalpractices in order to compete effectively for government contracts, and theyare in many cases required by the stipulations of the contract to abide byspecific financial and programmatic regulations (Smith forthcoming; Sosin,Smith, Hilton and Jordan forthcoming). Organizations with certain culturesor histories may resist these isomorphic tendencies, especially if governmentcontracts are a relatively small part of their overall revenue or if they pro-vide low-intensity services where government regulations are less intrusiveand extensive.

Overall, then, the governance of third-sector organizations changes quiteprofoundly as the contracting relationship evolves and develops, especiallyin terms of board governance, internal management practices, and externalrelations including political engagement. The politics of contracting encour-ages third-sector organizations to engage with government officials and build long-term relationships that position the organizations to obtain con-tracts and favorable policies and regulations. The ripple effect on the serviceprovider is internal restructuring to ensure that the organization reflects themanagement priorities of government.

Conclusion

Contracting with third-sector organizations has grown, owing in part to the influence of New Public Management thinking on the value of markets,greater responsiveness, and decentralization. Indeed, commissioners andpublic managers in the UK, the US and elsewhere are under increasing pres-sure to get the best-possible value for money when they are deciding howbest to commission and how far to involve the third sector in service designand delivery. Moreover, the increase in contracting for public services has brought a greater emphasis on performance and accountability. Yet, inorder to achieve better performance, government commissioners have been

pressured to depart from the classic market paradigm. In the UK, the central government is investing large sums of money in capacity-building and infrastructure development designed to improve the governance and management of the third sector. This same interest in capacity-building has been evident in many other countries including the US, Canada, NewZealand and Australia. Capacity-building has also promoted an investmentapproach to contracting that views the government–third-sector relationshipas a long-term proposition.

Whilst this capacity-building aims to improve governance, the simul-taneous increase in performance accountability and the development ofmore detailed regulation, inspection and quality assurance systems have inpractice removed from the boards of third-sector organizations the imme-diate need to take direct responsibility for service quality. Increasingly,boards tend to use the results of external inspections and benchmarking systems to judge their own performance, rather than themselves taking a directinterest in what is going on. Board agendas can often be consumed with thereception of inspection reports done by others.

The growth of contracting, pressures to achieve economies of scale, andthe sector’s responses to competition will likely mean that increasing pro-portions of the work go to a smaller number of larger contractors, especiallygiven the financial crisis affecting countries throughout the world. Largeregional and national third-sector providers find that community and service-user involvement is increasingly difficult to do; yet inspectors and regulatorsspend more time than before looking for evidence of service-user and community involvement.

In a very real sense, we are in the midst of an important transition period inthe third sector in the US, the UK and around the world as board memberslearn new behaviors individually and as groups, in response to wider changesin public policy pertaining to contracting and government regulation. How-ever, a paucity of evidence exists of the relationship between good governance,impact and outcomes, and more research needs to be done on this significanttopic across sectors. In addition, an urgent need exists perhaps to reclassifyservice providers so that the debate on quality and effectiveness can be movedaway from the goals and priorities of a core of traditional third-sector organizations such as large social service providers toward a debate aboutwhich organizational forms and governance styles are evidentially better toprovide really effective public services. This new debate would then enableobjective comparison between traditional third-sector agencies and newsocial enterprises and quasi-autonomous service providers (in the UK, forexample, city academies and foundation hospitals; in the US, charter schoolsand public development authorities) in a genuine attempt to learn which formsof governance really offer best value for public money. This reassessmentshould also include a review of the relative merits of small local community-based organizations, given their contribution to active civil society as well

as the implications for policy and practice of the growth of large providerorganizations. Inadvertently, current policies and practice on contracting maybe “killing the golden goose” by undermining effective performance and sustainability and community and civic engagement in third-sector organ-izations. Board members and staff of third-sector organizations, policy-makers, and advocates interested in improving the quality of services andstrengthening social citizenship should welcome this inquiry and debate.

Notes1 The authors would like to thank Putnam Barber, Mary Kay Gugerty, Margaret

Harris, and Stephen B. Page for comments on an earlier draft of this chapter and the financial support of the Nancy Bell Evans Center on Nonprofits andPhilanthropy at the Evans School of Public Affairs, the University of Washington,for research support in the preparation of this chapter.

2 Scotland and to an extent Wales have somewhat different policies.3 See, for example, www.in-control.org.uk4 In the US, the Bush administration, through its New Freedom Initiative (NFI),

offered individuals more choice in service providers and helped facilitate more clientindependence and work opportunities. NFI has financed a variety of demonstra-tion and pilot projects in support of these choice goals throughout the country.

5 For more information on Futurebuilders, see: http://www.futurebuilders-england.org.uk For information on Capacitybuilders, see: http://capacitybuilders.org.uk/

6 For example, the Department of Health Report praises the acquA (AcquiringAccreditation) initiative in Herefordshire as a good example of how third-sectororganizations can effectively organize and regulate themselves to meet the morerigorous demands of entering into contractual relationships with public-sectorproviders for the delivery of health and social care service (p. 19).

7 For information on US experience, see Berry 2003; Bass, Arons, Guinane, and Carter2007.

ReferencesAdams, Gina, Rohacek, Monice and Snyder, Kathleen (2008) Child Care Voucher

Programs: Provider Experiences in Five Countries, Washington, DC: The UrbanInstitute, http://www.urban.org/UploadedPDF/411667_provider_experiences.pdf

Alexander, Jennifer, Nank, Renee and Stivers, Camilla (1999) “Implications of Wel-fare Reform: Do Nonprofit Survival Strategies Threaten Civil Society”, Nonprofitand Voluntary Sector Quarterly, 28 (4): 452–75.

Allard, Scott W. (2008) Out of Reach: Place Poverty and the New American WelfareState, New Haven, Conn.: Yale University Press.

Alter, Kim (2007) Social Enterprise Typology, Portland, Oreg.: Virtue Ventures,http://www.virtueventures.com/setypology.pdf

Audit Commission (2008) Competition and Contestability: How Councils Can ImproveServices through Competition and Contestability, http://www.improvementnetwork.gov.uk/imp/core/page.do?pageId=1068494

Bass, Gary D., Arons, David F., Guinane, Kay and Carter, Matthew F. (2007) Seen but Heard: Strengthening Nonprofit Advocacy, Washington, DC: The AspenInstitute.

Behn, Robert D. (2001) Rethinking Democratic Accountability, Washington, DC: TheBrookings Institution.

Behn, Robert D. and Kant, Peter A. (1999) “Strategies for Avoiding the Pitfalls of Performance Contracting”, Public Productivity and Management Review, 22 (4):470–89.

Berry, Jeffrey M. and Arons, David (2003) A Voice for Nonprofits, Washington, DC:The Brookings Institution.

Better Regulation Task Force (BRTF) (2005) Better Regulation for Civil Society: MakingLife Easier for Those Who Help Others, http://archive.cabinetoffice.gov.uk/brc/upload/assets /www.brc.gov.uk/betregforcivil.pdf

Bode, Ingo (forthcoming) “Creeping Marketization and Post-corporatist Govern-ance: The Transformation of State–nonprofit Relations in Continental Europe”,in Susan Phillips and Steven Rathgeb Smith (eds) Governance and Regulation inthe Third Sector, Abingdon: Routledge.

Burger, Ary and Veldeer, Vic (2001) “The Growth of the Nonprofit Sector in theNetherlands”, Nonprofit and Voluntary Sector Quarterly, 30 (2): 221–46.

Cabinet Office (2006) Partnership in Public Services: An Action Plan for Third SectorInvolvement, http://www.cabinetoffice.gov.uk/media/cabinetoffice/third_sector/assets/psd_action_plan.pdf

Casey, John and Dalton, Bronwen (2006) “The Best of Times, the Worst of Times:Community-Sector Advocacy in the Age of ‘Compacts’ ”, Australian Journal ofPolitical Science, 41 (1): 23–38.

Charity Commission (2006) Partnership Strategy, http://www.charity-commission.gov.uk/spr/partner.asp

Compact (2008) The Concise Compact Guide, http://www.thecompact.org.uk/files/102396/FileName/TheConciseCompactGuide.pdf

Considine, Mark (2003) “Governance and Competition: The Role of Non-profitOrganisations in the Delivery of Public Services”, Australian Journal of PoliticalScience, 38 (1): 63–77.

Considine, Mark and Lewis, Jenny M. (2003) “Bureaucracy, Network, or Enter-prise? Comparing Models of Governance in Australia, Britain, the Netherlands,and New Zealand”, Public Administration Review, 63 (2): 131–40.

Deakin, Simon and Michie, Jonathan (1997) “Contracts and Competition: AnIntroduction”, Cambridge Journal of Economics, 21: 121–5.

Department of Children and Families (DCFS) (2006) Budget Holding Lead Professional (BLHP), www.dcsf.gov.uk/localauthorities/_documents/content/SCYPG07_14-attach.doc

Department of Health (2006a) Report of the Third Sector Commissioning Task Force – Part II Outputs and Implementation. http://www.dh.gov.uk/prod_consum_dh/groups/dh/groups/dh_digitalassets/@dh/@en/documents/digitalassets/dh_4137568.pdf

Department of Health (2006b) No Excuses. Embrace Partnership Now. Step towardschange! Report of the Third Sector Commissioning Task Force – Part Outputs andImplementation. http://www.dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/@dh/@en/documents/digitalasset/dh_4137568.pdf

Eikenberry, Angela M. and Kluver, Jodi Drapal (2004) “The Marketization of theNonprofit Sector: Civil Society at Risk?”, Public Administration Review, 64 (2):132–40.

Etherington, Stuart (2006) Annual Conference Keynote Speech, http://www.ncvo-vol.org.uk/press/speeches/?id=2465

Forsythe, Dall W. (ed.) (2001) Quicker, Better, Cheaper? Managing Performance inAmerican Government, Albany, NY: The Rockefeller Institute Press.

Foster, William and Bradach, Jeffrey (2005) “Should Nonprofits Seek Profits?”, HarvardBusiness Review, February: 92–100.

Friedman, Mark (2005) Trying Hard Is Not Enough: How to Produce MeasurableImprovements for Customers and Communities, Victoria, BC: Trafford.

Gillanders, Gillian (2008) The Budget Holding Lead Professional Pilots Have Lessons forAll Children’s Services Authorities, London: OPM, http://myblog.opm.co.uk/?p=57

Gronbjerg, Kirsten (1993) Understanding Nonprofit Funding, San Francisco, Calif.:Jossey-Bass.

Gutch, Richard (1992) Contracting Lessons from the US, London: National Councilof Voluntary Organisations.

Harris, Margaret and Rochester, Colin (eds) (2001) Voluntary Organisations and SocialPolicy in Britain, Basingstoke: Palgrave Macmillan.

Henriksen, Lars Skov and Bundesen, Peter (2004) “The Moving Frontier inDenmark: Voluntary–State Relationships since 1850”, Journal of Social Policy, 33 (4): 605–25.

Her Majesty’s Treasury (2002) The Role of the Voluntary and Community Sector in Service Delivery: A Cross Cutting Review, http://www.hm-treasury.gov.uk/d/CCRVolSec02.pdf

Hood, Christopher (1991) “A Public Management for All Seasons”, Public Admini-stration, 69 (1): 3–19.

Improvement and Development Agency (I&DeA) (2008) Peer Review: Helping YouImprove, http://www.idea.gov.uk/idk/aio/8748019

Independent Commission for Good Governance in Public Services (2004) The GoodGovernance Standard for Public Services, London: Office for Public Management(OPM)/Chartered Institute of Public Finance and Accountancy (CIPFA), http://www.opm.co.uk/resources/papers/policy/Good_Gov_Standard.pdf

Kettl, Donald (2005) The Global Public Management Revolution, Washington, DC:The Brookings Institution.

Kirkpatrick, Ian, Kitchener, Martin and Whipp, Richard (2001) “Out of Sight, Outof Mind: Assessing the Impact of Markets for Children’s Residential Care”,Public Administration, 79 (1): 49–71.

Lundstrom, Tommy and Svedberg, Lars (2003) “The Voluntary Sector in a SocialDemocratic State: The Case of Sweden”, Journal of Social Policy, 32 (2): 1–22.

Lyons, Mark (forthcoming) “Australia: A Continuing Love Affair with the New PublicManagement”, in Susan Phillips and Steven Rathgeb Smith (eds) Governance andRegulation in the Third Sector, Abingdon: Routledge.

Maryland Association of Nonprofits (2009) Standards for Excellence: An Ethics andAccountability Code for the Nonprofit Sector, http://www.marylandnonprofits.org/html/standards/documents/Booket507Revised.pdf

National Conference of Voluntary Organisations (NCVO) (2008) Governance HubStresses Need for Ongoing Collaboration and Investment in Governance Development,http://www.politics.co.uk/opinion-formers /press-releases/ncvo-governance-hub-stresses-need-ongoing-collaboration-and-investment-in-governance-development-$1215533$364415.htm

Office of Government Commerce (2006) Buy and Make a Difference: How to AddressSocial Issues in Procurement, http://www.ogc.gov.uk/documents/Social_Issues_in_Public_Procurement.pdf

Osborne, David and Gaebler, Ted (1993) Reinventing Government, New York: Plume.Panel on the Nonprofit Sector (2007) Principles of Good Governance and Ethical Practice,

Washington, DC: The Independent Sector, http://www.nonprofitpanel.org/report/principles/Principles_Guide.pdf

Phillips, Susan (forthcoming) “Regulating for Relational Governance of the Volun-tary Sector: Teetering toward Reform in Canada”, in Susan Phillips and StevenRathgeb Smith (eds) Governance and Regulation in the Third Sector, Abingdon:Routledge.

Prabhakar, Meera, Thom, Graham, Hurstfield, Jennifer and Parashar, Urvashi (2008)Individual Budgets for Families with Disabled Children, London: Department of Children and Families, http://www.dcsf.gov.uk/research/data/uploadfiles/DCSF-RR057.pdf

Priemus, Hugo (2000) “Housing Vouchers: A Contribution from Abroad”, in C. EugeneSteuerle et al. (eds) Vouchers and the Provision of Public Services, Washington, DC:The Brookings Institution.

Public Accounts Committee (2006) Working with the Voluntary Sector, http://www.publications.parliament.uk/pa/cm200506/cmselect /cmpubacc/717/717.pdf

Rhodes, R. A. W. (1996) “The New Governance: Governing without Government”,Political Studies, 44: 652–67.

Ring, Peter Smith and Van De Ven, Andrew H. (1992) “Structuring CooperativeRelationships between Organizations”, Strategic Management Journal, 13 (7): 483–98.

Salamon, Lester M. (1987) “Partners in Public Service: The Scope and Theory ofGovernment–Nonprofit Relations”, in Walter Powell (ed.) The Nonprofit Sector:A Research Handbook, New Haven, Conn.: Yale University Press.

Simonen, Leila and Kovalainen, Anne (1998) “Paradoxes of Social Care Restructur-ing: The Finnish Case”, in Jane Lewis (ed.) Gender, Social Care and Welfare StateRestructuring in Europe, Aldershot: Ashgate.

Smith, Steven Rathgeb (forthcoming) “The Government–Nonprofit Relationship inthe United States: New Challenges and Possibilities”, in Susan Phillips and StevenRathgeb Smith (eds) Governance and Regulation in the Third Sector, Abingdon:Routledge.

Smith, Steven Rathgeb and Lipsky, Michael (1993) Nonprofits for Hire: The WelfareState in the Age of Contracting, Cambridge, Mass.: Harvard University Press.

Smith, Steven Rathgeb and Smyth, Judith (1996) “Contracting for Services in aDecentralized System”, Journal of Public Administration Research and Theory, 6 (2): 277–96.

Sosin, Michael, Smith, Steven Rathgeb, Hilton, Timothy and Jordan, Lucy P.(forthcoming) “Temporary Crises and Priority Changes: The Case of State Sub-stance Abuse Systems”, Journal of Public Administration Research and Theory.

Van Slyke, David M. (2007) “Agents or Stewards: Using Theory to Understand theGovernment–Nonprofit Contracting Relationship”, Journal of Public AdministrationResearch and Theory, 17 (2): 157–88.

Part IV

GOVERNANCE OFINTERORGANIZATIONAL

NETWORKS

TRUST IN GOVERNANCENETWORKS: LOOKING FOR

CONDITIONS FOR INNOVATIVESOLUTIONS AND OUTCOMES

Erik-Hans Klijn

1. Introduction: governance, governance networks and trust

Authors writing on the subject of governance have begun to acknowledge itsnew buzzword status. However, an old adage reminds us that if a conceptis everything it is nothing (see also Frederickson 2005: 285). Frederickson(2005) asks what has become of public administration in recent years whenevery area of the field seems to have been subsumed under the broadumbrella of governance.1 This chapter seeks to discuss the value of trust ingovernance or governance networks. But, before we do so, we must first assessthe meaning of governance.

What is governance?

In his widely cited article, Rhodes (1997) provided six different inter-pretations of the word “governance”. His overview of governance coveredcorporate governance, new public management, good governance as a socio-cybernetic system, governance as a self-organizing network, and other aspects.Others have added to Rhodes’s definitions by including multi-level govern-ance and market governance (see Frederickson 2005; Bekkers et al. 2007).Looking more closely at all the interpretations, I see the following four majordefinitions to be dominant in the literature (Kooiman 1993; Rhodes 1997;Pierre and Peters 2000; Frederickson 2005; Osborne 2006; Sorensen and Torfing2007):

1. Governance as good governance or as corporate governance. In this view,governance refers to the principles of a properly functioning public adminis-tration. Such an administration is characterized by the fair treatment of

citizens and an unambiguous organization that adheres to the basic prin-ciples of the rule of law. The emphasis here is on the operation of government,rather than on the manner in which government is organized. However, this view of governance adds little to classical ideas of government or evenbureaucracy.

2. Governance as new public management, a means of improving per-formance and accountability or a form of market governance (Osborne and Gaebler 1992; Bekkers et al. 2007). Under this definition, the role ofgovernments is to steer rather than to row (Osborne and Gaebler 1992). The focus of government should be to set goals, and not to control the implementation process. Policy implementation is best left to separate public or private agencies which can be held accountable through the use of clear performance indicators and other market mechanisms. This defini-tion of governance is similar to that of new public management, which stressesthat governments should guide performance from a distance, using per-formance indicators and market mechanisms to arrange services and securepolicy outputs. Governance under this definition is almost akin to “politicalfunctioning”.

However, it is important to remember that governance is something completely different from the concept of New Public Management. Whilegovernance tends to emphasize the horizontal relationships between govern-mental organizations and other organizations, New Public Management can be considered an opposing paradigm to governance in many ways sinceit emphasizes central steering. In much of the New Public Management literature, one can find the assumption that the function of politics is to set clear goals, steer progress with clear and cleverly designed incentives, andthen leave implementation to other organizations (Osborne and Gaebler 1992; Hood 1991).

3. Governance as multi-level governance or intergovernmental relations.In some studies in the governance literature, governance is described as multi-layer government or intergovernmental governance. These are two separatestrands of literature that are distinctly different from each other. Nonethe-less, their common theme is the difficulty of achieving results in a multi-actorsetting. Although not all studies explicitly use the network concept, this literature stresses that networks are needed to address all aspects of govern-ance problems because these tend to cross public organization boundariesand their hierarchical levels. For example, governance issues such as thoserelated to the economic regeneration of deprived areas or environmental andpollution issues tend to cross a number of organizational and hierarchicalboundaries (Bache and Flinders 2004; Agranoff and McGuire 2003; Hoogeand Marks 2004), and the literature in this area focuses on specific types of networks in which public actors from various levels have prominent positions.

GOVERNANCE OF INTERORGANIZATIONAL NETWORKS

4. Governance as network governance (both self-steering or non-self-steering). Governance and the network concept are strongly related in someparts of the governance literature (Kooiman 1993; Rhodes 1997; Kickert et al. 1997; Sorensen and Torfing 2007). Governance takes place mainly within the fluid network of public and non-public actors, and the inter-action between these groups makes the processes of governance complex anddifficult to manage. Consequently, the steering and management strategiesrequired are different from that used in more classical approaches. The focus under this definition is on the complex interaction process, and the pro-cess of negotiation in a network of governmental organizations and otherorganizations that are both private and public (not-for-profit).

Looking at the four conceptualizations of governance, we can conclude thatthere is little reason to differentiate between governance and governance networks, since the two more interesting conceptualizations (3 and 4) of governance tend to stress that governance takes place within a networks of actors. At best, we can make a faint distinction by saying that govern-ance relates to the interaction process (and its guidance), while networks relate to the empirical phenomenon that policy issues are solved within networks of actors (for a more elaborate discussion of this distinction, see Klijn 2008). Thus, governance networks will be used here as an indi-cation of more or less stable patterns of social relationships (interactions, cognitions and rules) between mutually dependent public, semi-public and private actors that arise and build up around complex policy issues or policyprogrammes.2

Trust in governance networks: the crux of the issue

Trust is often mentioned as the core coordination mechanism of networks.Trust is often contrasted with two other forms of governance: markets andhierarchies (Thompson et al. 1991). Hierarchies are characterized by rules and central steering, while markets are dominated by the decentralizedmechanism of prices as dictated by supply and demand. The discussion ontrust as a core coordination mechanism in networks is often somewhat misleading and confusing in that it places networks somewhere between markets and hierarchies, when in fact this practice reveals little about net-works. This placement blurs more than it illuminates because networks tendto incorporate traces of several coordination mechanisms other than trustat the same time, including rules and central steering, market mechanismsand bargaining (see Koppenjan and Klijn 2004). Given this mix, the idea of trust as the core coordination mechanism in networks is not exceedinglyuseful. Nonetheless, we would do well to remember that the inclusion of trustis not entirely without merit.

TRUST IN GOVERNANCE NETWORKS

This chapter: the importance of trust in ensuring innovativegovernance network related outcomes

This chapter argues that trust is an important element of governance net-works. We shall demonstrate that the function of a governance network isessentially to facilitate negotiations and reconcile different values with eachother. For a network to be successful, innovative policy proposals have tobe achieved within the complex interaction process. Trust is an essential com-ponent that facilitates fluent interactions, the flow of information, and otherconditions necessary for actors to develop innovative solutions. However,trust does not arise automatically. It has to be consciously achieved throughimprovements in the interaction between actors and through dedicated net-work management activities (see also Huxham and Vangen 2005).

This chapter will first discuss the essence of governance networks, the valueof conflicts, and their necessity in fostering innovative policy proposals. It will then discuss the importance of trust and use empirical evidence to illustrate the contributions of trust to the process. The chapter concludes witha section on how to achieve trust in governance networks.

2. The essence of governance networks: political strugglebetween actors with differing values

To assess the value or importance of trust in governance networks, we mustfirst look at the essential characteristics of governance networks. It is rea-sonable to say that much of the literature on governance networks stressesthe complexity of the governance process, and emphasizes that these pro-cesses should provide new and innovative solutions for societal problems andimprove implementation (see Koppenjan and Klijn 2004; Skelcher et al. 2005; Sorenson and Torfing 2007). However, the policy problems that areaddressed within governance networks are also often characterized by valueconflicts and complex interaction and bargaining processes. So, although innovative policy solutions are necessary, they are not easily won since variousactors favor different, often conflicting values and policy solutions.

Governance networks as wicked issues, and value conflicts

Most of the issues that are dealt with within governance networks are fairlycomplex. Typical projects include large infrastructure developments, watersafety and water management projects, and other complex service deliveryinitiatives involving more than one partner. These problems are referred to in the literature as “wicked problems” (see Rittel and Webber 1973). Wicked problems are policy issues that involve many actors who often dis-agree about the nature of the problem and the desired solution. In addition,there is usually insufficient or controversial information surrounding these

problems, which makes it difficult to interpret them and to find appropriatesolutions. Wicked problems are difficult to resolve as they almost always involvetricky conflicts between values and scarce resources. Looking at decisionsrelated to infrastructure such as the building or expansion of a road, the ques-tion of which of the following values to prioritize arises in earnest:

• transport values (prioritizing the expansion of the road so that traffic isfacilitated)

• liveability values (prioritizing the values of inhabitants around the roadwho suffer from the negative external effects of its use and expansion)

• environmental values (prioritizing the elements of nature that will suffer as a result of the expansion of the road)

Such conflicting values make it difficult to derive a univocal criterion for good governance network outcomes (Kickert et al. 1997; Mandell 2001). In the above example, key actors who hold the differing values include environmental interest groups that promote environmental values, the ministry of transport that favors road expansion, and citizen/inhabitant groupswho hold liveability-related values that drive them to protect their living environments. All the actors demand a say in the decision-making process,and their inclusion makes decision-making within the governance networkimmensely complex.

However, while the process may be contentious, there is little that is illegitimate about it. None of the actors typically pursues unlawful efforts,and their attempts to influence the decision-making process can be seen asa normal part of a mature democracy in which actors can articulate theirinterests in a relatively open decision-making process. From this point of view,the often heard complaints in the world of politics, bureaucracy and the media about lengthy decision procedures, too many interest groups, and so on seem questionable. Fault-finders (usually ministries) often protestbecause they feel that their values are compromised in the trade-off againstthe values of other stakeholders.

Governance networks as complex interactions

Given the value difference, it is not surprising that a large number of actorsare part of the governance networks in which these wicked problems are dealt with. Recent research in the Netherlands shows that twelve differentorganizations are involved in an average environmental planning project (Klijn et al. 2010). It is typical for the network relationships to exist over alonger period of time, and to change as a result of the interactions betweenthe actors involved and their interpretations of the problem space (for a moreelaborate discussion of these ideas, see Koppenjan and Klijn 2004). Govern-ance networks thus manifest themselves in concrete policy interactions,

which can be seen as policy games that are played by the different actors.During these games, the actors attempt to influence policy issues, partly byreinterpreting the available information and the informal and formal rulesthat were previously generated.3 In short, governance networks can be char-acterized by intense dynamism and a high degree of complexity (Teisman et al. 2009), which makes them very difficult to manage. Managers and other parties involved in the negotiations are often surprised by the manyunexpected turns of events, and by the unexpected effects of the dynamicinteractions between individual actors and their strategic choices.

Governance networks as a conduit for the value-driven search for solutions

The wicked character of governance network processes renders them intrin-sically political in nature. Governance networks essentially present a strugglebetween differing values, how problems are defined, and how solutions arederived. If, following Easton (1953) (and many other Public Administra-tion experts), we define politics as the authoritative allocation of values, the political character of processes in networks becomes clear. However, italso becomes clear that this process of authoritative allocation no longer takes place solely within representational bodies. A number of writers haveargued that politics is removed from governance networks (e.g. Bovens et al. 1995) because politics has been dissolved, or at least partially dissolved,in the network itself where the struggle between all the different groups takesplace (also see Klijn and Skelcher 2007). All political institutions includingthe lower chamber, the various city councils, the provincial states and all the political parties are entrenched in the network of actors. They no longersimply realize the public interest from far above the turmoil of battle. Thus,the often-heard complaint that networks and the interest groups involved in them are a threat to democracy essentially implies that we (i.e. as membersof the media, citizens and academics) no longer recognize politics as a visible center of power.

There rarely is a cut-and-dried solution to many of the problems addressedthrough governance networks, no matter how eagerly the media make usbelieve that there is. Many of these issues involve problems of which we havelittle knowledge, problems that are ill-defined, and in which the basic informa-tion surrounding the issue is contested. Further, the problems addressed ingovernance networks rarely can be solved by quick unilateral measures. Thedifferent actors differ in their opinions about the desired solutions, and theseactors may include civilians, organized groups as well as governmental bodies.The decision-making process itself is a quest for solutions involving the collection of necessary information. Actors work to determine which solu-tions would be acceptable for as many of the parties involved. In this way,networks can be viewed as efforts to unite various values with one another.

Governance networks: the search for innovative solutions that combine values

The inter-active nature of the network process requires that a good attemptwas made to combine the values of various parties which requires new solutions. Creativity is essential to the generation of new solutions. Most ofthe solutions initially available for consideration are developed by partiesadvocating their own point of view and their own particular values. For thenetwork to be useful, it has to be able to generate new, innovative solutionsthat combine information arising from different actors and their resources.But the act of sharing information and resources, and of working cooper-atively toward an innovative solution, is often seen by individual actors orinterest groups as being risky. None of the actors can fully predict the formthat the solution will eventually take. Exchanging information with other actorscan lead to a situation in which other actors use the information for theirown benefit.

This situation presents a classic mix of the challenges associated with collective decision-making, strategic games and risk-taking (Axelrod 1984;Williamson 1996). Trust becomes even more essential for risk-taking whenthe actors have opposing perspectives on the nature of the problem, whatvalues are relevant, and the nature of a desirable solution. Governance net-works have the potential to serve as a viable means of developing trust and achieving interesting outcomes. This view of networks as a facilitatorof trust differs somewhat from the more commonly held view that trust isthe core coordination principle of networks, an argument that was criticizedat the start of this chapter.

3. Governance networks, trust and risk

In the context of the above discussion, one can begin to see networks as aviable means of achieving collective action. Such networks serve to solidifythe otherwise fluid interaction patterns between actors. Uncertainty and riskare key concepts in such governance networks. Thus, trust and managerialactivities can be expected to be very important because few authoritative mech-anisms exist that serve to bind actors, and the presence of multiple partiesheightens strategic uncertainty. What is said in the literature about trust, and its relation to risk and uncertainty, is interesting in this context, and soin this section we further examine the relationship between trust, risk anduncertainty. But, first, we need to establish a workable definition of trust.

What is trust?

Trust is defined in multiple ways in the literature. Before we start a dis-cussion about trust, we have to narrow the range. A few key characteristics

emerge from the literature (see Edelenbos and Klijn 2007). First, when anactor trusts another actor, he is in fact taking a risk, and allowing him-self to be vulnerable to opportunistic behavior. Each actor expects the otheractor to refrain from opportunistic behavior, even if the opportunity for it arises (Deakin and Michie 1997; Deakin and Wilkinson 1998). He or shetrusts that the partner will take their interests into account in the inter-action (Rousseau et al. 1998; Nooteboom 2002). In unpredictable situations,a minimum level of trust is a prerequisite for any pro-active undertaking(Gambetta 1988; Lane and Bachmann 1998). A conscious choice has to be made to take a risk, and this is usually done with the belief that the other party can be trusted.

A working definition for trust can thus be derived as follows: “Trust refers to the actors’ more or less stable, positive perception of the intentionsof other actors, that is, the perception that other actors will refrain from opportunistic behavior” (Edelenbos and Klijn 2007). Although trust can be recognized from the actions of the various actors, actions are in turn aresult of trust and we have to distinguish trust from the actions themselves.We also have to be careful to make a distinction between trust and institu-tional characteristics such as rules and norms which often serve to facilitatetrustworthy behaviors. These institutional features are not the same as trust. Only the more precise, limited definition of trust can be useful in research; and, as we shall see in section 4, it allows us to measure the levelof trust in governance networks by assessing perceptions of trust among network actors.

Trust and risk: a classic economic perspective

Trust does not play a role of any importance in most classical economic viewsof transactions and market relations. In fact, it is often seen as irrelevant toa normal market situation. Buyers and sellers are led by their own rationalbehavior, and simple transactions can be dealt with using classical contractsor a firm handshake. Most economic analysts see no use for the concept of trust, even in more complex situations involving specific investments andlong-term relations. Although most would accept that we face a situation of incomplete contracts (and incomplete information), they would suggestmeans other than trust to address these concerns. Most analysts would stressthat incomplete contracts should come with safeguard mechanisms to pro-tect the contractor against other parties. Their solution to the problem ofrisk would more likely lie in contractual safeguards and reputation-relatedincentives (see Lyons and Metha 1997; Lorenz 1999). These safeguards can take the form of bonds, penalties and other contractual agreements, andshould serve to prevent opportunistic behavior and facilitate the resolutionof ex-post disputes between partners. An alternative to devising safeguardsis vertical integration, which provides partners with an incentive to invest

efficiently and reduces transaction costs if transactions are frequent, specificand uncertain.

Another way to deal with the problem of uncertainty is to describe andanalyze it in terms of reputation: according to this view, actors have an interest in preserving a good reputation for the sake of future deals. Thisreputation argument is strongly inspired by game theory (Lorenz 1999) andis similar to the arguments of Axelrod (1984), who found that cooperationis possible in repeated prisoner’s dilemma games when the gains that can be earned in future (the “shadow of the future”) are great enough.

Williamson (1996), the godfather of transaction economics, goes as far as to tell us that trust is a confusing concept because it amounts to noth-ing more than risk-taking. He says that “calculative relations should bedescribed in calculative terms, to which the language of risk is exactly suited ”(Williamson 1996: 485–6). In his view, notions of trust blurs the argument,because they cause one party to accept the risk that the other party may beacting opportunistically. In his view, the notion of trust as acting in goodfaith (without calculation) does not add anything useful to the analysis ofthe situation.

Williamson distinguishes between trust at the institutional level (trust in the functioning of institutions such as the legal system) and trust at theindividual level. In Williamson’s neo-institutional perspective, risk-taking isreduced to the problem of calculating whether an organization will begin to work with another organization, and this can be deduced by asking: Isthe balance of benefits and costs favorable enough to take the risk that anothercontracting party might act opportunistically? Also, what risks can an organization afford to take? These questions apply also to concerns aboutreputation. They allow the problem to be reduced to one in which risks are analyzed and weighed.

Trust, risk and rationality: the need for trust

Most authors on trust agree that trust is inextricably related to risk. With-out risk, the notion of trust is simply unnecessary (Rousseau et al. 1998; Lyons and Metha 1997; Lane and Bachmann 1998; Nooteboom 2002). Asdemonstrated earlier, trust has much to do with the expectations an actorhas of another actor. In contractual relations, partnerships and almost all cooperative relations involving private and public actors of variousaffiliations, actors are confronted with risks that can take various forms. Oncewe recognize the various kinds of behavioral risks being faced, Williamson’sanalysis allows for an immediate clarification of the problem. Under this analysis, trust is seen as almost unnecessary, and the concept is viewed asbeing almost synonymous with risk-taking. The risk-taking perspectiveassumes that a calculation can be made which then forms the basis for arational decision. Without such a calculation, the assumption of risk becomes

just as blind as trusting the other actors. However, this assumption cannoteasily be squared with the idea of bounded rationality, which is also the starting point in the neo-institutional approach. If the possibility of gatheringinformation is limited, then so are the opportunities to assess the behavioralalternatives that independent actors have. Interestingly, this argument holdseven more strongly for larger governance networks with greater numbers of actors and where unexpected strategic moves take place more frequentlyand the possibilities are wider. The uncertainty over the strategic behaviorof actors is thus much greater than the term “information insecurity” sug-gests, certainly as it is presented in the literature on governance networks.In her attempts to define away the social context of choices, Williamson’sneo-institutional economics thus implicitly assumes the hyper-rationalism in her argument against trust – a rationalism for which she often critiquesclassic economic approaches.

Trust and governance networks: a perfect couple

The value of the concept of trust lies in the fact that the environment is too complex for us to foresee all the possible contingencies, reason them out or calculate them accurately (Deakin and Wilkinson 1998). It is preciselythe many possible contingencies that present themselves which makes the concept of trust such an interesting one to consider when appraising the performance of governance networks (Edelenbos and Klijn 2007).

Trust provides an anchor with which actors in governance networksdetermine what risks actors will or will not take in cooperative relations, and it allows them to make their choices accordingly. The fact that trust is not the same as calculation and cannot be entirely reduced to purely rational assessments does not mean, as Williamson suggests, that it is blind.Trust does in fact have a rational basis. Most authors point to the fact that trust grows according to the actor’s earlier experiences (Nooteboom et al. 1997; Lyons and Metha 1997; Rousseau et al. 1998) and that trust will not be sustained when it is repeatedly violated (Rousseau et al. 1998;Nooteboom 2002) as evidence that the decision to trust is usually well considered. Reciprocal behavior is an essential condition for trust to emergeand to be sustained. This reciprocity is precisely what happens in governancenetworks.

In other words, trust can only exist by virtue of dependency (Rousseau et al. 1998) and earlier interactions, and these are core features of any established governance network. Actors need each other to realize the out-comes they cannot achieve alone, and networks bring about intensifying interactions and dependencies. Interestingly, these are exactly the conditionsmentioned in the literature on trust. One could therefore see networks as avehicle for achieving trust, rather than assuming (as is done in some of theliterature) that trust is an inherent characteristic of networks.

4. The value of trust in governance networks: theoretical and empirical considerations

Why would trust be a favorable condition for the functioning of governancenetworks and the achievement of positive outcomes in governance net-works? Although not inherently written from a governance point of view,several reasons can be found in the literature on trust (see also Edelenbosand Klijn 2007):

• trust reduces transaction costs• trust facilitates cooperation and stability in (network) relations• trust stimulates learning and knowledge exchange• trust stimulates innovation

We shall look at each of these arguments in turn and examine some empirical evidence that supports them.

Trust reduces transaction costs

Trust may bring about a saving in transaction costs. In his book on trust,Fukuyama argues: “Property rights, contracts, and commercial law are allindispensable institutions for creating a modern market-oriented economicsystem, but it is possible to economize substantially on transaction costs ifsuch institutions are supplemented by social capital and trust” (Fukuyama1995: 336). Because trust brings about greater predictability in the beha-vior of key actors, it serves to reduce the risks inherent in transactions andimprove the cooperative relationship. After all, in a situation where one actor assumes good intentions on the part of the other, the likelihood of opportunistic behavior leading to unexpected interactions is smaller. But trustalso serves as an alternative to contracts, or as a welcome addition to them.Writing contracts is associated with a transaction cost (Williamson 1996),and the more complex the interaction, and the more complex the desired outcomes or products, the more effort actors have to put in to draw a goodcontract. Not surprisingly, we do not find many contracts and formalizedorganizational forms emerging from governance networks (see Klijn 2009for an elaboration of how most public–private partnerships can be seen asrelative loosely coupled organizational forms). Tight contracts and heavy organ-izational forms do not match very well with the complexity and dynamismof the interactions and the decision-making that occurs in governance networks.

As earlier mentioned, in this environment, trust becomes a welcome addi-tion to, or even replacement for contracts (Sako 1998; Ring and Van de Ven1992; Nooteboom 2002). It reduces the need to include extensive clauses inthe contract that cover possibilities that may arise in the future.

Trust enhances investment and stability in relations

The classical neo-economic perspective does not yet fully embrace the con-cept of trust. Instead it focuses very keenly on the cost aspect and remainsrather static in its search for explanations, concentrating almost entirely onproduction costs, transaction costs, organizational costs and costs resultingfrom opportunistic behavior in the context of the development of a new tech-nology. Williamson acknowledges (see Nooteboom 2002) that, even thoughthe study of long-term interactions is far more complex, the classical per-spective maintains its focus on one-off interactions. It pays little attentionto profit, or the relational or investment dynamics (Ring and Van de Ven1992; Nooteboom et al. 1996). In contrast, in the field of governance net-works, long-term interaction is a large area of debate along with dependencyand the creation of new innovative products or policy solutions. A greaterlevel of attention is being paid to potential revenues than to costs, and tocontinuous interaction over the long term instead of to single transactions(see Sako 1998; or Parker and Vaidya 2001).

Trust increases the probability that actors will invest (with resources like money, knowledge, etc.) in cooperation, and it creates stability in therelationship. If trust is present, actors are more willing to invest despite the risks involved. Following the argument of competitive advantages in theprivate sector (Sako 1998; Nooteboom 2002), one could say that govern-ance networks with a higher level of trust between the actors gain a “policyadvantage” in that the lower levels of distrust and uncertainty and thestronger ties between players empower the actors to move more quickly andforcefully.

Trust stimulates learning and knowledge exchange

Much of the knowledge in modern networks, such as that related to spe-cific routines and contacts, is tacit. It lies within specific individuals or organizations (Parker and Vaidya 2001; Nooteboom 2002) and is acquiredover many years of experience. It can only be accessed through frequentexchanges and more intensive cooperation. A similar observation can be maderegarding the importance of learning (Lundvall 1993). Learning and discoveringnew things require a high level of knowledge exchange and intensive inter-action. Because actors lack the resources and knowledge to realize solutions,they have to interact to acquire them. The performance of these networksis probably strongly dependent on the extent to which they succeed in thisexchange.

Needless to say, these forms of knowledge exchange require a minimumof trust to take place. Drawing up a contract in such a network of organiza-tions is far too costly and may even be impossible, given the limited meansand typically limited access that parties provide to information that they may

hold (Graeber 1993; Parker and Vaidya 2001). Trust serves to facilitate andencourage the flow of information between the actors in governance networks(Edelenbos and Klijn 2007).

Trust stimulates innovation in governance networks

Last but not least is the argument that governance networks need to createinnovative solutions for complex problems. As we know from the litera-ture on research and development (R&D) partnerships, the problem with innovation is that it is as risky as it is desirable. R&D has the potential to give firms a clear competitive advantage, but it is impossible to ascertainbeforehand if something valuable will come out of the R&D efforts, or ifthe other partner will do his or her share to further the most desirable outcome (Parker and Vaidya 2001; Lundvall 1993).

From a transaction cost perspective, vertical integration is optimal for thissituation (Williamson 1996); but this, too, has its disadvantages. Vertical integration provides incentives for players to minimize their differences, butthis has a negative effect on innovation. It is fair to say that the cognitivedistance between partners should be neither so great that it prohibits cooperation nor so small that it prohibits innovation (Nooteboom 2000). Itis these opposing forces that help explain why organizations in the privatesphere increasingly choose to interact within networks (Miles and Snow 1986;Alter and Hage 1993; Lundvalt 1993; Parker and Vaidya 2001).

One of the main reasons for the emergence of governance networks is thefact that they have to deal with wicked problems that require the coopera-tion of many different actors who have to develop innovative solutions together.Trust is thus invaluable in governance networks as it facilitates innovationby reducing uncertainty about opportunistic behavior, and encourages thehorizontal exchange of information and cooperation. One would predict thatnetworks that demonstrate higher levels of trust would show more innova-tive results and perform better in solving complex policy problems.

Does trust matter in governance networks? Empirical evidence

Although there is a substantial body of research being conducted on matters of trust, this area is seldom addressed in the field of public adminis-tration and is almost completely absent in the governance literature (for exceptions, see Edelenbos and Klijn 2007; Klijn et al. 2010). From researchdone in business administration, especially on partnerships, we know thathigher levels of trust have several positive effects, including better relationswith suppliers (Nooteboom 2002), which translate into lower costs. Trust alsoenables relational governance between actors and facilitates cooperation.Examining the literature to determine whether trust matters, McEvily andZaheer concluded that it does: “In most optimistic sense, we find a number

of studies have produced results that are consistent with our own findingsthat trust enhances inter-organizational performance” (McEvily and Zaheer2006: 294). But does this also hold for governance networks?

More recently, Klijn, Edelenbos and Steijn (2010) have looked at the relationship between trust and outcomes in governance networks. Judgingfrom their survey of actors involved in a large environmental project, thereappears to be a high correlation between the level of trust (as measured byfive items) and perceived outcomes. Respondents to this survey made a clear distinction between process and content outcomes, and the researchersfound in their regression analysis (which controlled for actor characteristics[e.g. public or private], phase of decision-making and the complexity of the issue) that trust has a very strong relationship with both content (Betaof .534) and process outcomes (Beta of .545). Thus, we are inclined to con-clude that trust does matter in governance networks.

5. No trust without (network) managing efforts

It is clear that trust does not arise unaided. If one looks at the many case studies that have been written on governance networks, it is clear thatinteractions are dominated by actors following their own goals. There is typically a fair amount of distrust related to conflicting strategies, leadingto deadlock and other negative outcomes (see Mandell 2001; Marcussen and Torfing 2007). These problems demonstrate a lack of trust and maketwo additional things clear:

1. Trust is not a characteristic which is naturally present in networks;2. To achieve trust, it is necessary to facilitate intensive levels of inter-

action, and the network has to be actively managed.

Trust does not arise unaided: the need for network management

Despite the critical nature of trust, it cannot be assumed to be present inevery governance network. The research actually suggests that trust is a relatively scarce commodity in many governance networks and has to be developed over time (Zucker 1986; Lane and Bachmann 1998; Edelenbos andKlijn 2007) by intensifying the interaction between actors. But the greaterinteraction will only lead to greater trust if the actors see real benefits arisingfrom the interaction (see Nooteboom 2002).

The intensification of the interaction in the pursuit of the development of interesting and rewarding content is a crucial managerial task that formsa key part of the governance and supervision of any governance network.Looking at the literature, network management strategies fall into four maincategories (see Agranoff and McGuire 2001; Mandell 2001; Koppenjan andKlijn 2004):

• arranging: this includes strategies to organize the interactions in govern-ance networks into temporary organizational structures

• exploring content: exploring the varying views of different actors, the newsolutions that may arise from interactions with them, and connecting theideas of different actors

• connecting: securing contacts between actors, improving relations, etc.• process agreements: agreements about process rules and methods of

interaction between the actors

We know from a wide range of case studies that employing network managerial strategies improves the outcomes of networks (see, for instance,Marcussen and Torfing 2007). In particular, the studies of Meier andO’Toole (2001, 2007) on the Texas educational districts have large samplesizes that convincingly demonstrate the impact of network management strategies, along with other studies (see also Huang and Provan 2007; Klijnet al. 2010). But do network management strategies really enhance trust? What is the nature of the relationship between trust outcomes and networkmanagement in governance networks?

The impact of network management on trust

There are very few empirical studies of governance networks that exam-ine the relationship between trust and network management strategies.Although several studies suggest that increasing the intensity of the inter-actions and employing network management strategies enhance the relationsbetween actors in the governance network (see, for instance, Marcussen andTorfing 2007; Huang and Provan 2007), the level of trust has not been directlyrelated to the intensity and character of the network management strategiesemployed.

Klijn, Edelenbos and Steijn (2010) asked actors involved in environ-mental projects questions about their network management strategies. Theyconstructed sixteen items that measured the four types of network manage-ment strategies mentioned above, and found a fairly strong correlationbetween the number of strategies employed in governance networks and the overall level of trust. Interestingly, when the outcomes were included in the regression analysis, it was found that the relation between trust andoutcomes diminishes but still remains significant. There also appears to be a significant relationship between network management strategies and outcomes.

So it can be said that both trust and the number of employed networkmanagement strategies have independent effects on network outcomes, butnetwork management also positively connects with trust. It seems that betterresults are achieved in governance networks if active network managementis combined with and stimulates a higher level of trust.

6. Conclusion: the value of trust in governance networks

There has been little said about the role of trust in governance networks other than that it serves as a key coordinating mechanism. However, as we have seen in this chapter, this statement is highly debatable since net-works are characterized by many coordinating mechanisms. The theoreticalliterature in fields other than that of governance networks, particularly public administration, provides ample support for the view, as does the scarce empirical material available on the subject.

The main reason for trust’s positive impact on the performance of net-works lies in its beneficial effect on the cooperation between the actors inthe networks and the flow of information it facilitates between actors for thepurposes of innovation. Since networks are mostly formed around wickedproblems that require such innovative solutions along with the resources of multiple actors, the beneficial effects of trust are clear. Trust also seemsto be a promising concept for further examination in the discussion on governance and governance networks.

1 If one were to seek articles with the word “governance” in Scopus, over 8,000 articles are found. The articles fall under the category of social science, with a large number being from public administration and environmental planning. Asignificant rise is seen in the number of governance-related between 1996 (106) and2007 (1193).

2 Despite the stability in the relations of the network, interactions within networkscan still be capricious, because actors deploy their own strategies. For an elabor-ation of this point, see Koppenjan and Klijn 2004.

3 Governance networks can thus be seen as the temporarily solidified form of thesepolicy games: the pattern of interactions at a certain moment in time, the set ofperceptions of the actors (and the resemblances and differences) at that momentand the set of valid informal and formal rules at that moment.

References

Agranoff, R. and McGuire, M. (2001) “Big Questions in Public Network Manage-ment Research”, Journal of Public Administration Research and Theory, 11: 295–326.

Agranoff, R. and McGuire, M. (2003) Collaborative Public Management: New Stra-tegies for Local Governments, Washington, DC: Georgetown University Press.

Alter, C. and Hage, J. (1993) Organizations Working Together, Newbury Park,Calif.: Sage.

Axelrod, R. (1984) The Evolution of Cooperation, New York: Basic Books.Bache, I. and Flinders, M. (eds) (2004) Multilevel Governance, Oxford: Oxford

University Press.Bekkers, V., Dijkstra, G., Edwards, A. and Fenger, M. (2007) Governance and the

Democratic Deficit: Assessing the Democratic Legitimacy of Governance Practices,Aldershot: Ashgate.

Bovens, M. et al. (1995) De verplaatsing van de politiek, Amsterdam.Deakin, S. and Michie, J. (eds) (1997) Contract, Co-operation, and Competition: Studies

in Economics, Management and Law, Oxford: Oxford University Press.Deakin, S. and Wilkinson, F. (1998) “Contract Law and the Economics of Inter-

organizational Trust”, in C. R. Lane and R. Bachmann (eds) Trust Within andBetween Organizations: Conceptual Issues and Empirical Applications, Oxford:Oxford University Press.

Easton, D. (1953) A Systems Analysis of Political Life, New York: John Wiley.Edelenbos, J. and Klijn, E. H. (2007) “Trust in Complex Decision-making

Networks: A Theoretical and Empirical Exploration”, Administration and Society,39 (1): 25–50.

Ferlie, E., Lynn, L. E. and Pollitt, C. (eds) (2005) The Oxford Handbook of PublicManagement, Oxford: Oxford University Press.

Frederickson, H. G. (2005) “What Happened to Public Administration? Govern-ance, Governance Everywhere”, in E. Ferlie, L. E. Lynn and C. Pollitt (eds) The Oxford Handbook of Public Management, Oxford: Oxford University Press.

Fukuyama, F. (1995) Trust: The Social Virtues and the Creation of Prosperity, NewYork: The Free Press.

Gambetta, D. (1988) Trust: Making and Breaking of Cooperative Relations, Oxford:Blackwell.

Graeber G. (ed.) (1993) The Embedded Firm: Understanding Networks. Actors,Resources and Processes in Interfirm Cooperation, London: Routledge.

Hood, C. (1991) “A Public Management for All Seasons”, Public Administration, 69 (Spring): 3–19.

Huang, K. and Provan, K. G. (2007) “Structural Embeddedness and OrganizationalSocial Outcomes in a Centrally Governed Mental Health Service Network”,Public Management Review, 9 (2): 169–89.

Huxham, C. and Vangen, S. (2005) Managing to Collaborate: The Theory andPractice of Collaborative Advantage, Abingdon: Routledge.

Kickert, W. J. M., Klijn, E. H. and Koppenjan, J. F. M. (1997) Managing ComplexNetworks: Strategies for the Public Sector, London: Sage.

Klijn, E. H. (2008) “Governance and Governance Networks in Europe: An Assess-ment of 10 Years of Research on the Theme”, Public Management Review, 10 (4):505–25.

Klijn, E. H. (2009) Public Private Partnerships in The Netherlands: policy, projectsand lessons, Economic Affairs, March 2009: 26–32.

Klijn, E. H., Steijn, B., Edelenbos, J. (2010) The impact of network management strategies on the outcomes in governance networks, in Public Administration 2010(forthcoming).

Klijn, E. H., Edelenbos, J. and Steijn, B. (forthcoming, 2010a) Trust in governancenetworks; its impact on outcomes, Administration and Society (forthcoming 2010a).

Klijn, E. H. and Skelcher, C. K. (2007) “Democracy and Governance Networks:Compatible or Not? Four Conjectures and Their Implications”, Public Admini-stration, 85.

Kooiman, J. (ed.) (1993) Modern Governance: New Government–Society Interactions,Newbury Park, Calif.: Sage.

Koppenjan, J. F. M. and Klijn, E. H. (2004) Managing Uncertainties in Networks: ANetwork Approach to Problem Solving and Decision Making, Abingdon: Routledge.

Lane, C. and Bachmann, R. (eds) (1998) Trust Within and Between Organizations:Conceptual Issues and Empirical Applications, Oxford: Oxford University Press.

Lorenz, E. (1999) “Trust Contract and Economic Cooperation”, Cambridge Journalof Economics, 23: 301–15.

Lundvall, B. A. (1993) “Explaining Interfirm Cooperation: Limits of the Transaction-cost Approach”, in G. Graeber (ed.) The Embedded Firm: Understanding Networks.Actors, Resources and Processes in Interfirm Cooperation, London: Routledge.

Lyons, B. and Metha, J. (1997) “Private Sector Business Contracts: The Text betweenthe Lines”, in S. Deakin and J. Michie (eds) (1997) Contract, Co-operation, andCompetition: Studies in Economics, Management and Law, Oxford: Oxford Univer-sity Press.

Mandell, M. P. (ed.) (2001) Getting Results through Collaboration, Westport, Conn.:Quorum.

Marcussen, M. and Torfing, J. (eds) (2007) Democratic Network Governance inEurope, Cheltenham: Edward Elgar.

Marks, G. and Hooghe, G. L. (2004) “Contrasting Views on Multi-level Govern-ance”, in I. Bache, and M. Flinders (eds) (2004) Multilevel Governance, Oxford:Oxford University Press.

McEvily, B., Zaheer, A. (2006) Does trust still matter? Research on the role of trustin inter-organisational exchange, in: R. Bachmann, A. Zaheer (eds) 2006, Hand-book of Trust Research, Cheltenham: Edward Elgar Publishing.

Meier, K. J. and O’Toole, L. J. (2001) “Managerial Strategies and Behaviour inNetworks: A Model with Evidence from US Public Education”, Journal of PublicAdministration and Theory, 11 (3): 271–93.

Meier, K. J. and O’Toole, L. J. (2007) “Modelling Public Management: EmpiricalAnalysis of the Management–performance Nexus”, Public Administration Review,9 (4): 503–27.

Miles, R. E. and Snow, C. C. (1986) “Organization: New Concepts for New Forms”,California Management Review, 28 (3).

Nooteboom, B. (2002) Trust: Forms, Foundations, Functions, Failures and Figures,Cheltenham: Edgar Elgar.

Nooteboom, B., Berger, H. and Noorderhaven, N. (1997) “Effects of Trust andGovernance on Relational Risk”, Academy of Management Journal, 40 (2): 308–38.

Osborne, D. and Gaebler, T. (1992) Reinventing Government: How the Entre-preneurial Spirit Is Transforming the Public Sector, Reading, Mass.: Addison-Wesley.

Osborne, S. P. (2006) “The New Public Governance”, Public Management Review,8 (3): 377–87.

Parker, D. and Vaidya, K. (2001) “An Economic Perspective on InnovationNetworks”, in O. Jones, S. Conway and F. Steward (eds) Social Interaction and Organisatonal Change: Aston Perspectives on Innovation Networks, London:Imperial College Press.

Pierre, J. and Peters, B. Guy (2000) Governance, Politics and the State, Basingstoke:Macmillan.

Ring, P. S., and Van der Ven, A. (1992) “Structuring Cooperative Relations betweenOrganizations”, Strategic Management Journal, 13: 483–98.

Rittel, H. J. W. and Webber, M. M. (1973) “Dilemmas in a General Theory ofPlanning”, Policy Sciences, 4: 155–69.

Rousseau, D., Sitkin, S. B., Burt, R. S. and Camerer, C. (1998) “Not So DifferentAfter All: A Cross-discipline View of Trust”, Academy of Management Review, 23 (3): 393– 404.

Sako, M. (1998) “Does Trust Improve Business Performance?”, in C. Lane and R. Bachmann (eds) Trust Within and Between Organizations: Conceptual Issues andEmpirical Applications, Oxford: Oxford University Press.

Skelcher, C., Mathur, N. and Smith, M. (2005) “The Public Governance ofCollaborative Spaces: Discourse, Design and Democracy”, Public Administration,83 (3): 573–96.

Sorensen E. and Torfing, J. (eds) (2007) Theories of Democratic Network Govern-ance, Cheltenham: Edward Elgar.

Teisman, G. R., van Buuren, A. and Gerrits, L. (eds) (2009) Managing ComplexGovernance Networks, Abingdon: Routledge.

Thompson, G. J. Frances, Levacic, R. and Mitchel, J. (1991) Markets, Hierarchiesand Networks: The Coordination of Social Life, London: Sage.

Williamson, O. E. (1996) The Mechanisms of Governance, Oxford: Oxford UniversityPress.

Zucker, L. (1986) “Production of Trust: Institutional Sources of Economic Structure,1840–1920”, Research in Organizational Behavior, 8: 53–111.

IMPLEMENTATION ANDMANAGERIAL NETWORKING INTHE NEW PUBLIC GOVERNANCE

Laurence J. O’Toole, Jr and Kenneth J. Meier

Abstract

Implementing public programs and delivering public services in the era of New Public Governance call for public managers to interact externallyand often collaboratively with a range of stakeholders. Such relationships,it is often argued, build support for implementation, buffer programs fromunexpected and negative shocks, and induce co-productive contributions from potential partners during execution. This chapter presents empirical evidence from hundreds of public organizations about how such networkingbehavior, which may constitute a form of organizational social capital,influences outcomes. At the same time, internal management within hierar-chies also seems to matter for results. One of the key aspects of the researchagenda on the New Public Governance is to sort through these relationshipsfor their performance-relevant implications.

The era of the New Public Governance signals considerable promise as wellas substantial challenges. The latticing of horizontal linkages across actorsand institutions, overlaying and potentially complicating the more familiarvertical ones, requires careful analysis. Networks of policy-involved actorsoften work in complex ways to shape policy choice and action (Bardach 1998;Huxham 2000; Stoker 2004; Bingham and O’Leary 2008). In some nations,corporatist arrangements and assumptions bind disparate “social partners”in processes of decision-making. Even in much less corporatist and consider-ably more pluralist systems, arrangements like Public–Private Partnershipsand government contracting – relational and otherwise – are now com-monplace. The increasing attention directed to many countries’ commitmentto sustainable development induces ever more intricate interweaving of policy sectors and social actors to address the cross-cutting governanceneeds impelled by this daunting policy agenda (Lafferty 2004). For reasons

having to do with political motives, technical requirements for successful implementation, and the increasingly ambitious policy agendas of govern-ments, webs of intertwined organizations and other institutions routinely co-produce outputs and outcomes. Successfully achieving policy results insuch governance contexts requires that these complicated multi-actor insti-tutional settings be managed effectively.

This chapter overviews the challenge of policy implementation in complexgovernance settings involving networks of interdependent actors. We first framethe implementation theme and establish the importance of its multi-actor,networked character in contemporary governance. We then review brieflysome of what is known about the determinants of implementation success orfailure. We concentrate in particular on one key driver of implementationperformance – the actions of public managers, particularly as they interactin and with the set of interdependent actors in other organizations. In thisregard we distill some of the recent research findings that bear on public management and policy implementation in complex settings, and sketch some related questions deserving of further research attention.

Policy implementation in complex settings

When considering “policy implementation” we are interested in what happensbetween the establishment of a governmental intention to do something –or stop doing something – and the consequent effect of that decision in theworld of action. Emphasizing an era of governance implies that imple-mentation action involves a multiplicity of actors contributing to the resultsof public programs, rather than merely the “lonely organization” (Hjern 1982)assumed in some early studies.

But is that so? Anecdotal and case-study evidence is abundant, and systematic data are increasingly available as well. The answer from thegrowing number of the latter is clear: networks of interdependent organ-izations – or parts of organizations – are typically involved in trying to turnpolicy goals into reality. Local public managers in the UK (Walker, O’Tooleand Meier 2007) and the US (Agranoff and McGuire 2003) routinely reportextensive connections with external actors in their environment. Swedish Public Employment Service offices and municipalities regularly interact, andpolicy implementation is assisted when both trust and goal congruence arepresent (Lundin 2007). Local networks contribute to the implementation ofagricultural watershed management and the adoption of environmental best-management practices (Lubell and Fulton 2008). Dutch institutions of highereducation sit in networks the characteristics of which help shape performance(Schalk Torenvlied and Allen forthcoming). Hundreds of cases of inter-organizational collaboration in numerous policy fields have been identifiedat the subnational level in Thailand (Krueathep, Riccucci and Suwanmalaforthcoming). The great majority of new or substantially revised national

IMPLEMENTATION AND MANAGERIAL NETWORKING

public programs in the US require or strongly encourage participation bymultiple organizations, governments, and often sectors in the US (Hall andO’Toole 2000, 2004). And a recent meta-analysis of 137 cases of collabora-tive governance including evidence from multiple countries and policy fieldssuggests regularities in what makes for successful operation (Ansell and Gash2008). Networked, collaborative and interorganizational implementationsettings are clearly the rule rather than the exception.

Does it matter? A number of theoretical arguments provide reasons to thinkso (for instance, Kickert, Klijn and Koppenjan 1997; Provan and Milward2001; Agranoff and McGuire 2003; see also Rhodes 1997; Stoker 1999).Certainly inducements to and constraints inhibiting cooperation across organ-izational boundaries in networked and collaborative settings are different from– and typically more challenging than – generating successful cooperativeimplementation via unified organizations. Authority is typically weaker ininterorganizational situations, and encouragement toward cooperation mustperforce rely less on established communication channels, routines andshared worldviews than it does within hierarchical agencies. One import-ant implication is that public management itself is likely to be even moreimportant to successful implementation in networks than in hierarchies(O’Toole 2000a), and “management” in these cases will surely involve con-siderable negotiation, framing, activation (and deactivation) of connectionsamong nodes, and trust-building across organizational lines.

Determinants of implementation success in networkedsettings: the role of the manager

The huge literature on policy implementation in complex multi-actor settingsincludes many efforts to identify key variables that shape results but oftenrelatively little in the way of parsimonious theoretical exposition (for reviewsof the research literature, see O’Toole 1986, 2000b, 2004). Exacerbating theproblem has been a case-study literature that cannot in principle provide muchin the way of a valid, multivariate exposition. Nevertheless, some progress hasrecently been made, particularly with regard to an often underemphasizedcontributor to implementation success in networks: public management itself.

In part to remedy this gap, we initiated a research program more than a decade ago systematically to model and explore the relationship betweenpublic management and public program performance during implementation,particularly in networked settings. After reviewing the case-study literatureon managers, management, and their putative role in executing policy, weformalized some of the key ideas available into a testable model (O’Tooleand Meier 1999).1

Our model incorporates three basic principles with regard to public man-agement and administrative systems for delivering public program results.First, such arrays are autoregressive systems – that is, they create processes

and operating procedures that tend to reproduce the same outputs over time.Second, the model is complex and nonlinear rather than strictly additive. Attimes variables interact in a multiplicative manner, at times the interactionis with a reciprocal function, and at other times terms add together and theninteract with another variable to generate their overall impact on public pro-gram performance. Third, the model is contingent to reflect our view thatwhat works in terms of public management is contingent on a variety of otherfactors. Among the most interesting contingencies are those involving net-works, managerial networking, and the environment.

The model contains three different functions of management. They are efforts to manage the internal operations of the organization,2 efforts to exploitopportunities in the interdependent environment,3 and efforts to limit the negative impact of environmental perturbations on the administrative sys-tem.4 The latter two functions in the second, or environmental, portion of the model are often combined.5 Our particular focus in this chapter is thismanagerial function of operating outward toward and in the interdependentenvironment of the organization – the ability to manage in the network (thus,managerial networking). We also devote some brief attention to aspects ofthe first-listed managerial function, internal management in the organization,including the managerial effort devoted to handling impacts from the inter-dependent environment that penetrate the core organization, particularly indifficult situations.

The theory of public management and program performance that we have been developing is highly parsimonious; it contains only four variables,or variable clusters: performance, management, stability, and the environ-ment. Recognizing that an operationalization and full testing of this modelwith any existing dataset is impossible, we have opted for an incrementalstrategy that focuses on the link between management and performance ingeneral – and specifically on developing reliable and valid measures of man-agement. Within this strategic approach, we have proceeded by testing discreteportions of the model and building on the results in subsequent rounds ofanalysis. In doing so, we have always controlled for the other two variableclusters, either statistically6 or by holding a variable relatively constantacross cases.7 In most cases we have conducted these studies in a pooled time-series analysis of Texas school districts.8 Unless otherwise noted, theempirical findings are generated from this database (for a scientific justifica-tion of the extensive use of this database, see Meier and O’Toole 2007).

Managers, networking and networks: evidence from a research program

Public managers coordinate and support production-oriented action insidetheir organizations, and they also work outward toward and in the inter-dependent environment as they network to build support and enhance

implementation results. The latter efforts may involve managers’ build-ing support for program implementation among interested stakeholders, facilitating collaborative efforts at program execution with other involvedorganizations, protecting core production from threatening aspects of the organ-izational environment, linking or disconnecting network actors from eachother to improve policy execution, or even using persuasion and bargainingskill to change the nature of the interdependent co-produced action. Suchefforts by managers, especially when executed with skill, may constitute a form of organizational social capital that can be important for imple-mentation in today’s policy settings.

We have conducted extensive empirical study of this networking functionand how it is related to policy outcomes. Our measure of managerial net-working assumes that managers cannot engage in network-like behavior withother actors in the environment without coming into contact with them. Usingthe Texas School District dataset, we asked top managers to rate how fre-quently, from daily to never, they interact with each of a set of environmentalactors (five actors in a 2000 survey, eight actors in 2002 and 2005 surveys,and ten in the 2007 survey). These items have been factor-analyzed and consistently produce a first factor that is a general networking measure withall positive loadings (at times one factor only is produced). This factor scoreis used as the measure of managerial networking.

In support of the networking measure

Does this measure tap managerial networking activity in a valid and reli-able way? The evidence is rather strong that it does. The factor analysis of networking items shows a consistent pattern across nodes; the networkedcontacts are all correlated with each other, and analysis always produces ageneric first factor with positive loading regardless of how many nodes areincluded in the analysis.9

The networking measure is also positively correlated with a manager’s estimate of how much of her effort is directed externally rather than focusedon matters internal to the organization (Meier and O’Toole 2003). The mea-sure also shows links to performance as tapped with numerous indicators(O’Toole and Meier 2003b: 54, 56), thus demonstrating substantial empiricalsupport and external validity.

Furthermore, a comparison of networking data between 2000 and 2002,as well as comparisons for later surveys such as 2005 and 2007, show thatnetworking by top managers is very much a managerial choice rather thana behavior pattern forced upon managers by external actors. Networking measures for the top manager in a given organization at the two time pointswere essentially uncorrelated if the organization had changed managers butwere strongly correlated when the same manager was in place in both timepoints (Meier and O’Toole 2005). Goerdel (2006) has also shown that when

the manager reports pro-active (self-initiated) interactions with externalnodes the link between networking and performance is more strongly posi-tive. All these findings support the use of this measure as a way of tappingmanagers’ activities in the networked environment in studies of interorgan-izational implementation and the role of managers in that process.

Is managerial networking related to implementation performance?

Estimating the relationship between managerial networking and program out-comes requires controlling for the other influences, particularly constraints andresources, that can also shape results. Accordingly, we used several years ofdata across all Texas school districts for which we were able to gather surveydata from top managers10 and have controlled for several resources and con-straints that vary over time and across jurisdictions. Meier and O’Toole (2001:285) and later papers showed that managerial networking was positively relatedto student performance on standardized tests and on high-end indicators forcollege-bound students. The relationship held even when controlling for pastperformance – a particularly tough test. Interviews with managers indicatedthat higher networking activity often signaled that managers were address-ing problems in the organizational environment, particularly political ones,and keeping such distractions from affecting the organization’s productionpersonnel. In this sense, public education managers’ networking activities arelikely directed toward building interorganizational routines or trust amongnetwork members. A more detailed assessment of contents of the reportednetworking by managers is a research subject that still awaits exploration.

Clearly, considerable additional research is needed if we are to understandfully the ways that managerial networking matters for implementation resultsin various places and fields. Some of that work has begun and extends thepattern reported here for school districts to local law enforcement. Nicholson-Crotty and O’Toole (2004) created an external management scale that combined measures of contact with measures of public feedback systems andwith community-oriented policing activities. They found a strong positive rela-tionship between this measure of networking and environmental managementand performance measured as crime clearance rates.

How much difference does managerial networking make? The maximumeffect size is estimated at approximately 5 to 6 percent of performance.11 Inother words, managerial networking does not turn implementation disasterinto brilliant success, but it makes an important and measurable difference– particularly given that other variables likely to be important in generatingresults, like task difficulty or financial resources, are controlled for in theseanalyses.

The nonlinear aspects of the relationship between managerial networkingand performance have been explored in two articles (Meier and O’Toole 2001,

2003; see also Hicklin, O’Toole and Meier 2008). An interesting relationshipwas found in the link between managerial networking and the autoregres-sive character of the administrative system. Inertia can obviously help whenimplementation is going well, but excessive inertia indicates a rigid system;such systems have difficulty adapting to change and therefore are likely to suffer implementation deficits over time. Meier and O’Toole (2003: 696)found that at high levels of managerial networking (more than 1.5 standarddeviations above the mean), the degree of autoregressivity declines rapidly.This relationship suggests that managing in the networked setting can pro-vide a public organization with some flexibility and allow it to break awayfrom suboptimal routines when useful. We also found greater efforts to man-age externally were associated with more support from program clientele (Meierand O’Toole 2003: 693).

The second, or environmental, term of the model referenced earlier suggeststhat managerial networking interacts with various resources and constraintsin a nonlinear fashion. Theoretically speaking, the notion here is that undercertain conditions management should enhance the positive impact ofresources considerably above normal or be able to mitigate the performance-dampening effect of constraints. Meier and O’Toole (2001: 289) showed thatmanagerial networking can indeed reduce the negative impact of black andHispanic students (measures of task difficulty regarding more disadvantagedstudents) as well as of noncertified teachers on performance, as expected bythe theory. (On the different forms of buffering that management may beable to employ to support the core production in a public program, see O’Tooleand Meier 2003a; empirical studies are reported in Meier and O’Toole 2009;Meier, O’Toole and Hicklin forthcoming). Similarly, managerial network-ing selectively interacts with some resources and can produce much largergains for a unit increase in resources in organizations with high levels of managerial networking (Meier and O’Toole 2003: 696).

Managerial networking also interacts with program performance in non-linear ways. When the data were split into quintiles based on the level oforganizational performance, the relationship between management and per-formance varied across the five subgroups. Managerial networking mattersmore for units at the extremes of performance – for organizations in the highest and lowest performing quintiles (Meier and O’Toole 2001: 695). Theformer case is likely the more important one, since almost anything man-agement is likely to do will improve the sorry performance of organizationsat the bottom. When implementation performance is quite high, managershave the option of deploying their networking in pursuit of more, or morecomplicated, opportunities in the interdependent environment, thus leveragingthe impact of networking more than usual.

What about the possibility of diminishing or even negative marginalreturns from networking by managers? Hicklin, O’Toole and Meier (2008)investigated whether nonlinearity applied to managerial networking taken

by itself. The analysis showed just such a pattern and also demonstrated that talented managers (measured as those scoring high on a measure of managerial quality: see below) seemed aware of the limits of networking, sincethey restricted their own efforts before they experienced negative marginalreturns. In sum, networking efforts by managers provide a positive contri-bution to performance, nonlinearities can be documented, and such networkingbehavior is subject to diminishing returns.

Networks and networking: differentiated performance impacts

Much of our empirical work has focused on the behavior of public managersand the impact of what they do on the implementation of public programs.Managerial networking, for instance, is one important such behavior. Butnetworks themselves are structural entities; we define them as patterns of twoor more units, in which not all major components are encompassed withina single hierarchical array (O’Toole 1997).

We know that structural properties of implementation networks are likelyto help shape policy outcomes, but estimating such impacts is difficult formethodological reasons. Detailed studies of network structures are labor-intensive, typically involve one or a few cases, and thus cannot isolate theeffects of network-structural properties on policy results (for a small-N effortto make the connection, see Provan and Milward 1995). In addition, maxi-mizing the variance in terms of network characteristics typically meansexamining cases from different policy fields. Since comparable performancemetrics rarely span different sectors of policy, the difficulties involved in probing the network-structure-to-results linkage are even more imposing.

In an effort to explore both behavioral networking and structural prop-erties of networks, we estimated the effects of managerial networking in moreand also less networked structural settings – with the structural feature includedin the analysis an aspect of the network defined in financial terms – the degreeto which the school district had to depend on other levels of government forbudgetary support (state and federal governments). Managerial networkingis positively linked to performance both when the organization is financiallydependent on others and when it is financially independent, but it is moreimportant for performance among organizations that are dependent on others for financial support. In short, managerial networking matters morein structural networks (O’Toole and Meier 2004b: 487–8).

Networking: quantity and quality

Considerable evidence indicates that managerial networking can boost publicprogram performance, and that more networking – to a point – generatesmore results. But does the quality of managerial efforts make a differenceas well? The answer is yes.

We have developed a general measure of managerial quality and appliedit to the Texas schools dataset. The measure does not isolate the quality ofnetworking alone but rather the quality of management overall. The networkingmeasure clearly taps only quantity of activity; the quality measure includesthe quality dimension of both internal and external management. Incor-porating such a measure of managerial quality in a large-N analysis presentsdaunting obstacles. We responded by developing a measure that incor-porates the judgments of locally informed political overseers, the individualschool boards, in annual decisions about remuneration for top managers.The top managers in question are relatively mobile individuals in a competitivelabor market with extensive information on salaries and performance. In such a situation, the annual salary determination by a school board can beexpected to include some assessment of the quality of job the manager hasdone. The measure essentially predicts what the manager’s salary should be,based on a set of factors that are highly correlated with salary (district size,human capital investments, prior performance, etc.), and uses the differencebetween predicted salaries and actual salaries as an estimate of managementquality (see Meier and O’Toole 2002 for details).

Analysis shows that top management quality is related to implementationresults for almost all performance indicators (Meier and O’Toole 2002). And,since the quality measure and the networking-quantity measure are uncor-related, their combined effect size is the sum of the two taken separately; for the public entities in question, this is approximately double that for thenetworking behavior alone.

Do all clientele win from managerial networking?

Much of the literature on networks and networking – linked as it often isto themes of co-production and collaboration – says or at least implies thatwhen managers stimulate interorganizational linkages everyone wins. The literature certainly recognizes that sometimes it is difficult to make these networked connections, but networking per se is typically portrayed as a potentially important but also fundamentally apolitical or technocratic enter-prise (for more detailed coverage, see O’Toole and Meier 2004a). But studiesin administrative politics, including some venerable classics (Selznick 1949),suggest a political dimension to managers’ efforts to interact with, attend to, and build bridges with external actors. We examined this question acrosshundreds of school districts. Because actors in the interdependent environmentthat are engaged in exchange relationships with a core public organizationare likely to over-represent the more advantaged parties in any social setting, we explored the distributional aspect of managerial networking byseeing who benefits from this form of managerial behavior.

We found that networking was positively related to higher test scores for Anglo students (as well as overall test scores) and three measures of

performance for college-bound students. However, managerial networkingwas unrelated to pass rates for Latinos, blacks, or low-income students(O’Toole and Meier 2004a: 688).12 This pattern indicates the operation of apolitics of public management whereby the more powerful political forcesare likely to exert greater influence on management as managers increase their activities in the network. For those interested in policy implementationin interorganizational settings, accordingly, the distributional dimension of performance should be a subject of sustained attention. Operating in net-works and engaging in networking constitute activities that do not suspendthe patterns of exchange and influence in the broader system; they likely obscuresuch channels but may even amplify their impacts.

Implementation in an era of the New Public Governance: concluding notes

The foregoing coverage, drawn largely from an extended research programon the role of public management in the execution of public programs andthe delivery of policy results, clearly demonstrates the importance of inter-organizational patterns, and especially the networking behavior of managers,in shaping implementation results. Considerable advancement has beenmade. Two cautionary points should nonetheless be noted. First, for all theprogress made, we do not yet know nearly enough about how networks and networking shape performance. And, second, even in a time of a NewPublic Governance, the internal management of public organizations – andperhaps the relationship between internally and externally oriented management– requires serious attention.

We know that networking matters, and also that structural features of networks have performance impacts as well. Furthermore, cross-national comparative analysis with large datasets has been initiated and should paysubstantial dividends. But understanding implementation and its manage-ment in patterns resembling the New Public Governance requires additionaladvances and other lines of research.

The research program we have developed for the past decade offerssignificant findings on a host of issues important to implementation and performance, but it has also had its limits. In particular, while we have analyzed managerial behavior outward into the networked environment, wehave not conducted empirical work on the management of networks them-selves. There are substantial difficulties in doing so (see O’Toole 2000a fordetails), but the topic deserves attention. We have developed some initial theoretical ideas and sought to formalize them (Meier and O’Toole 2004),but the management of networks for implementation merits substantially more attention.

In addition, we have extended our research program beyond the network-related aspects of public management to explore the ways that internal

management also contributes to policy outcomes. While a careful review of the findings would extend far beyond the reach of this chapter’s theme,it is important to note that several aspects of internal management contributepositively to implementation results – and do so independently of the exter-nally oriented aspects of management. For example, the management anddevelopment of a public organization’s human capital is strongly related to performance across a wide array of outcomes (O’Toole and Meier 2009). Human resource management has also been shown to be related tothe performance of local law enforcement agencies (Nicholson-Crotty andO’Toole 2004).

Internal management consists of much more than the management of humanresources, of course. Other aspects of inward-directed management have also been systematically analyzed, with similarly positive effects on programresults. Our studies of crisis management, whether from a sizable budget-cut or from natural disasters, show that managers make adjustments in the allocation of financial resources and staff to minimize disruptions in theimplementation of their core activities (Meier, O’Toole and Hicklin forth-coming; Meier and O’Toole 2009). Management capacity, which can sometimesseem like wasted or slack resources, can contribute to results – including orespecially during times of organizational stress.

Given these findings, it would be wrongheaded to proclaim an end to hierarchy or to the importance of its effective management for implementa-tion (Olsen 2006). Indeed, our findings regarding the importance of internalmanagement and of hierarchy are generally supportive of the evidencereviewed in hundreds of studies by Hill and Lynn (2005). Accordingly, it islikely that management of hierarchies, as well as of externally oriented, inter-organizational management, typifies implementation in today’s governancesystems. The usual characterization of the “New Public Governance” beingdeveloped by observers refers to undeniably important forces, but plenty of the more venerable influences remain. Understanding the combinedinfluences of the more complex patterns now visible should be a researchpriority for the field.

Acknowledgment

This chapter is part of an ongoing research agenda on the role of public management in complex policy settings. That agenda has benefited from the helpful comments of George Boyne, Stuart Bretschneider, Gene Brewer,John Bryson, Amy Kneedler Donahue, Sergio Fernández, H. GeorgeFrederickson, Carolyn Heinrich, Peter Hupe, Patricia Ingraham, J. EdwardKellough, H. Brinton Milward, Sean Nicholson-Crotty, David Peterson, Hal G. Rainey, Bob Stein and Vicky M. Wilkins on various aspects of thisresearch program. Needless to say, this article is the responsibility of the authors only.

Notes1 The model itself, which aims to depict the relationship among public manage-

ment, institutional arrangements, and public program performance, is as follows:Ot = b1(S + M1)Ot−1 + b2(Xt/S)(M3/M4) + et where O is some measure of outcome;S is a measure of stability; M denotes management, which can be divided intothree parts: M1 management’s contribution to organizational stability through additions to hierarchy/structure as well as regular operations, M3 management’sefforts to exploit the environment of the organization, M4 management’s effortto buffer the unit from environmental shocks; X is a vector of environmental forces; e is an error term; the other subscripts denote time periods; and b1 and b2

are estimable parameters. In the model, S can be considered a composite of thevarious kinds of stability in an organizational setting. Stability means constancyin the design, functioning and direction of an administrative system over time.Some stability is induced by structure; other forms of stability have their originsin management processes and procedures. A discussion of these factors can befound in O’Toole and Meier (1999) and the various empirical works cited in thischapter.

2 This aspect of management is the M1 function in the formal model depicted innote 1.

3 This managerial aspect is depicted as M3 in the formal model.4 This aspect appears as M4 in the formal model.5 The combination of these two functions can be formally represented by M2, defined

as the ratio of M3 to M4. The M2 term can thus replace that ratio in a slightlysimplified version of the formal model.

6 As with the “X” vector of resources and constraints from the environment.7 As with the “S” term, since we have often conducted our large-N studies across

structurally similar public organizations.8 School districts in the United States are generally independent local govern-

ments with their own taxing powers. All districts in the analyses discussed in thischapter are of this type. “Independent” means that the school district is not sub-ordinate to another unit such as a city. Independent districts have their own electedboard, have the ability to tax and set budgets, and acquire bonding authority by a vote of the residents. There are roughly 14,000 school districts in the US,approximately 1,000 of which are in Texas.

9 With the school-district dataset, we have factor analyzed reported networking data involving varying numbers of nodes. The factor scores are very highly correlated with each other for a given set of survey results, a finding strongly supportive of the measure’s validity and reliability. Analysis of similar survey datagathered from local-authority managers in the United Kingdom also consistentlyyields a first factor with a set of positive loadings (Walker, O’Toole, and Meier2007: 750).

10 We have used time series of varying lengths, most typically four or five years,and we have surveyed several times: 2000, 2002, 2004–5, and 2007. For certain research questions, we have used up to twenty years of performance data.

11 The full maximum effect from what managers do is considerably larger than this amount. The result reported here omits, for example, effects from internalmanagement. It also includes only the quantity of networking, not necessarily thequality; see below for some treatment of the latter. This estimate does not includeany of the nonlinear impacts (see below).

12 Networking was associated with lower dropout rates, a disadvantaged indicator,but was not related to attendance rates, another low-end indicator.

References

Agranoff, Robert and McGuire, Michael (2003) Collaborative Public Management:New Strategies for Local Governments, Washington, DC: Georgetown UniversityPress.

Ansell, Chris and Gash, Alison (2008) “Collaborative Governance in Theory and Practice”, Journal of Public Administration Research and Theory, 18 (4):543–71.

Bardach, Eugene (1998) Getting Agencies to Work Together, Washington, DC: TheBrookings Institution.

Bingham, Lisa Blomgren and O’Leary, Rosemary (2008) Big Ideas in CollaborativePublic Management, Armonk, NY: M. E. Sharpe.

Goerdel, Holly T. (2006) “Taking Initiative: Proactive Management and Organiza-tional Performance in Networked Environments”, Journal of Public AdministrationResearch and Theory, 16 (3): 351–67.

Hall, Thad E. and O’Toole, Jr, Laurence J. (2000) “Structures for PolicyImplementation: An Analysis of National Legislation, 1965–66 and 1993–94”,Administration and Society, 31 (6): 667–86.

Hall, Thad E. and O’Toole, Jr, Laurence J. (2004) “Shaping Formal Networks throughthe Regulatory Process”, Administration and Society, 36 (2): 1–22.

Hicklin, Alisa, O’Toole, Jr, Laurence, J. and Mier, Kenneth J. (2008) “Serpents inthe Sand: Managerial Networking and Nonlinear Influences on OrganizationalPerformance”, Journal of Public Administration Research and Theory, 18 (2):253–73.

Hill, Carolyn J. and Lynn, Jr, Laurence E. (2005) “Is Hierarchical Governance inDecline? Evidence from Empirical Research”, Journal of Public AdministrationResearch and Theory, 15 (2): 173–95.

Hjern, Benny (1982) “Implementation Research – the Link Gone Missing”, Journalof Public Policy, 2: 301–8.

Huxham, C. (2000) “The Challenge of Collaborative Governance”, Public Manage-ment Review, 2: 337–52.

Kickert, Walter J. M., Klijn, Erik-Hans and Koppenjan, Joop F. M. (1997)Managing Complex Networks: Strategies for the Public Sector, London: Sage.

Krueathep, Weerasak, Riccucci, Norma M. and Suwanmala, Charas (forthcoming)“Why Do Agencies Work Together? The Determinants of Network Formation atthe Subnational Level of Government in Thailand”, Journal of Public Adminis-tration Research and Theory.

Lafferty, William (2004) Governance for Sustainable Development: The Challenge ofAdapting Form to Function, Cheltenham: Edward Elgar.

Lubell, Mark and Fulton, Allan (2008) “Local Policy Networks and AgriculturalWatershed Management”, Journal of Public Administration Research and Theory,18 (4): 673–96.

Lundin, Martin (2007) “Explaining Cooperation: How Resource Interdependence,Goal Congruence, and Trust Affect Joint Actions in Policy Implementation”, Journalof Public Administration Research and Theory, 17 (4): 651–72.

Meier, Kenneth J. and O’Toole, Jr, Laurence J. (2001) “Managerial Strategies andBehavior in Networks: A Model with Evidence from US Public Education”,Journal of Public Administration Research and Theory, 11 (3): 271–95.

Meier, Kenneth J. and O’Toole, Jr, Laurence J. (2002) “Public Management andOrganizational Performance: The Impact of Managerial Quality”, Journal ofPolicy Analysis and Management, 21: 629–43.

Meier, Kenneth J. and O’Toole, Jr, Laurence J. (2003) “Public Management andEducational Performance: The Impact of Managerial Networking”, PublicAdministration Review, 63: 675–85.

Meier, Kenneth J. and O’Toole, Jr, Laurence J. (2004) “Conceptual Issues in Mode-ling and Measuring Management and Its Impacts on Performance”, in PatriciaIngraham and Laurence E. Lynn, Jr (eds) The Art of Governance: Analyzing Man-agement and Administration, Washington, DC: Georgetown University Press.

Meier, Kenneth J. and O’Toole, Jr, Laurence J. (2005) “Managerial Networking: Issuesof Measurement and Research Design”, Administration and Society, 37: 523–41.

Meier, Kenneth J. and O’Toole, Jr, Laurence J. (2007) “Modeling Public Manage-ment: Empirical Analysis of the Management–Performance Nexus”, PublicManagement Review, 9: 503–27.

Meier, Kenneth J. and O’Toole, Jr, Laurence J. (2009) “The Dog That Didn’t Bark:How Public Managers Handle Environmental Shocks”, Public Administration, 87 (3): 485–502.

Meier, Kenneth J., O’Toole, Jr, Laurence J. and Hicklin, Alisa K. (forthcoming) “I’veSeen Fire and I’ve Seen Rain: Public Management and Performance after aNatural Disaster”, Administration and Society.

Nicholson-Crotty, Sean and O’Toole, Jr, Laurence J. (2004) “Public Managementand Organizational Performance: The Case of Law Enforcement Agencies”, Journalof Public Administration Research and Theory, 14 (1): 1–18.

Olsen, Johan P. (2006) “Maybe It Is Time to Rediscover Bureaucracy”, Journal ofPublic Administration Research and Theory, 16 (1): 1–24.

O’Toole, Laurence J., Jr (1986) “Policy Recommendations for Multi-actor Imple-mentation: An Assessment of the Field”, Journal of Public Policy, 6 (2): 181–210.

O’Toole, Laurence J., Jr (1997) “Treating Networks Seriously: Practical and Research-based Agendas in Public Administration”, Public Administration Review, 57 (1):45–52.

O’Toole, Laurence J., Jr (2000a) “Different Public Managements? Implications ofStructural Context in Hierarchies and Networks”, in Jeffrey Brudney, Laurence J.O’Toole, Jr, and Hal G. Rainey (eds) Advancing Public Management, Washington,DC: Georgetown University Press.

O’Toole, Laurence J., Jr (2000b) “Research on Policy Implementation: Assessmentand Prospect”, Journal of Public Administration Research and Theory, 10 (2): 263–88.

O’Toole, Laurence J., Jr (2004) “The Theory–practice Issue in Policy Imple-mentation Research”, Public Administration, 82 (2): 309–29.

O’Toole, Laurence J., Jr and Meier, Kenneth J. (1999) “Modeling the Impact of PublicManagement: Implications of Structural Context”, Journal of Public Administra-tion Research and Theory, 9 (4): 505–26.

O’Toole, Laurence J., Jr and Meier, Kenneth J. (2003a) “Bureaucracy and Uncer-tainty”, in Barry C. Burden (ed.) Uncertainty in American Politics, New York:Cambridge University Press.

O’Toole, Laurence J., Jr and Meier, Kenneth J. (2003b) “Plus ça change: PublicManagement, Personnel Stability, and Organizational Performance”, Journal of PublicAdministration Research and Theory, 13: 43–64.

O’Toole, Laurence J., Jr and Meier, Kenneth J. (2004a) “Desperately SeekingSelznick: Cooptation and the Dark Side of Public Management in Networks”, PublicAdministration Review, 64 (6): 681–93.

O’Toole, Laurence J., Jr and Meier, Kenneth J. (2004b) “Public Management inIntergovernmental Networks: Matching Structural Networks and ManagerialNetworking”, Journal of Public Administration Research and Theory, 14: 469–95.

O’Toole, Laurence J., Jr and Meier, Kenneth J. (2009) “The Human Side of PublicOrganizations: Internal Management and Organizational Performance”, AmericanReview of Public Administration, 39 (5): 499–518.

Provan, Keith G. and Milward, H. Brinton (1995) “A Preliminary Theory of Inter-organizational Network Effectiveness”, Administrative Science Quarterly, 40 (1):1–33.

Provan, Keith G. and Milward, H. Brinton (2001) “Do Networks Really Work? A Framework for Evaluating Public Sector Organizational Networks”, PublicAdministration Review, 61 (4): 414–23.

Rhodes, R. A. W. (1997) Understanding Governance: Policy Networks, Reflexivity and Accountability, Buckingham: Open University Press.

Schalk, Jelmer, Torenvlied, René, Allen, Jim (forthcoming) “Network Embedded-ness and Organizational Performance: The Strength of Strong Ties in DutchHigher Education”, Journal of Public Administration Research and Theory.

Selznick, Philip (1949) TVA and the Grass Roots: A Study in the Sociology of FormalOrganization, Berkeley, Calif.: University of California Press.

Stoker, G. (ed.) (1999) The New Management of British Local Governance, Basing-stoke: Palgrave Macmillan.

Stoker, G. (2004) Modernizing British Local Government: From Thatcherism to NewLabour, Basingstoke: Palgrave Macmillan.

Walker, Richard M., O’Toole, Jr, Laurence J. and Meier, Kenneth J. (2007) “It’sWhere You Are That Matters: An Empirical Analysis of the Networking Beha-viour of English Local Government Officers”, Public Administration, 85: 739–56.

FROM NEW PUBLICMANAGEMENT TO

NETWORKED COMMUNITYGOVERNANCE? STRATEGIC

LOCAL PUBLIC SERVICENETWORKS IN ENGLAND

Steve Martin

Introduction

This chapter examines the impact on local public service networks in Englandof a range of policies which are inspired by the concept of networked community governance. This new(ish) approach is predicated on the beliefthat local government need not be directly involved in the delivery of local services but that it does have a key role to play in orchestrating partnershipsof local providers from across the public, private and voluntary sectors. Thechapter first describes the traditional model of the local welfare state whichheld sway in the period immediately following World War II. Next it con-siders the policies pursued by the Thatcher governments of the 1980s andearly 1990s which are widely seen as having been influenced by and symp-tomatic of the “New Public Management” (NPM). It then focuses on recentpolicies and their impact on local government and local governance. It arguesthat the increasing influence of networked community governance reflects an awareness of some of the weaknesses of previous approaches. However,it has not entirely displaced previous paradigms. Current policies blendapproaches associated with traditional models of public administration, theNPM and networked community governance. They are not therefore entirelynew but they have led to an increased emphasis on collaboration and “citizen-centered” services. This requires local officials and politicians to workacross organizational boundaries in ways that were not expected of them inthe past, and it poses challenges to traditional forms of central governmentoversight of local government and other service providers.

The local welfare state

In contrast to most other Western European countries, in the UK the rolesand responsibilities of local government are not codified, and there are no constitutional guarantees of its existence. The national parliament is sovereign and is able to create, restructure or abolish other tiers of govern-ment at will. Local government’s position is further weakened by its lack of financial autonomy. Ministries often regard local authorities simply as thedelivery agents of national policies rather than as democratically elected entities with their own independent mandate and locally determined priorities.And the absence of constitutional checks and balances means that it is very easy for the executive to enact new legislation and implement organ-izational changes (Pollitt 2007), with the result that local authorities frequentlycomplain of “initiative overload”.

British local authorities have no direct responsibility for the delivery ofhealthcare, but they play a major role in the provision of most other localpublic services. In the period immediately following World War II, local government was seen as the main vehicle for the delivery of the newly createdwelfare state and in particular in the provision of state-funded education,social care and housing. The 1944 Education Act established free com-pulsory schooling from age 5 to 15. Central government set the overall policyframework, but within broad parameters county councils were free to decidehow education was delivered locally. Some local education authorities (LEAs)chose to provide primary schooling until the age of 11, whilst others movedchildren from primary to middle schools at the age of 9 and then on intosecondary education at 14. Some separated the most academically gifted at the age of 11 and sent them to specialist (grammar) schools. Others hadnonselective systems and sent all children to mixed-ability (comprehen-sive) schools. Local government also oversaw post-16 and a wide range ofsocial services including child protection, services for vulnerable adults (forexample, ex-offenders and people with mental health problems, learning and/orphysical disabilities) and personal care for older people. Local authoritieshoused a significant proportion of the population. The 1957 Housing Actplaced a duty on them to evaluate housing conditions in their area and provide additional homes where necessary, and by the late 1970s councilsowned almost 7 million dwellings (Wilson and Game 2006). They were alsoresponsible for a range of public utilities as well as for consumer protection,environmental health, fire and rescue services, emergency planning, highwaysmaintenance, transport and traffic management, leisure, arts, recreation,libraries, planning, and the collection and disposal of waste.

Councils were therefore monopoly suppliers of key local public services.They had large budgets and were major employers. For the most part, they were organized around services. Directors of education, social services,housing and other key functions ran specialist departments which often had

little contact with other parts of the organization. Their activities were over-seen by committees of local politicians who, like the professional staff, oftenbecame specialists in particular services. Users of services (clients, pupils andtenants) had no choice of who provided services and little direct influenceover how they were run.

The fragmented local state

In the late 1960s and throughout the 1970s the combination of spiraling costs of welfare provision and a succession of economic crises began to callinto question the postwar model of welfare provision. The assumption thatpublic services should be provided directly by local authorities no longer seemedsustainable. The Secretary of State for the Environment famously informedcouncils that “for the time being at least, the party is over”. Their budgetscame under increasing pressure. The number of lower-tier (district) councilswas reduced, and on the recommendation of the Bains Committee, whichwas established by central government to advise on the internal managementof local authorities, many councils underwent significant internal restructur-ing. Chief executives, senior management teams, and policy and resourcescommittees were encouraged to assert corporate priorities, and in the pro-cess began to erode the power of the large, functionally organized servicedepartments that had been such a dominant feature of local government inthe immediate postwar period.

The Conservative government that came to power in 1979 believed thatBritain’s future economic prosperity depended on gaining control of publicspending and breaking the power of large public-sector trade unions. In pursuit of these twin objectives, it embarked on a series of policies whichwere inspired by the NPM and set it on a collision course with many localauthorities. Hood (1991) identifies seven “doctrines” of the NPM: “hands-on professional management”; explicit standards and measures of performance;greater emphasis on output controls; the disaggregation of units; greater com-petition; the promotion of private-sector management practice; an emphasison greater discipline and parsimony in resource use. Pollitt (1995: 133) offersa similar list of “core elements” of the NPM which he suggests “comprise akind of ‘shopping basket’ for those who wish to modernize the public sectorof Western industrial societies”. He adds to Hood’s list the introduction intopublic-sector organizations of more flexible working (including short-termcontracts, performance-related pay and local bargaining) and an emphasison service quality and responsiveness to service users. Operating at a slightlyhigher level of abstraction, Pollitt and Bouckaert (2000) identify four broadstrategies for public services reform: the maintenance of controls on publicspending; modernization of the administrative system through the introduc-tion of private-sector management practices and reform of political and managerial structures and practices; marketization of public services through

STRATEGIC LOCAL PUBLIC SERVICE NETWORKS IN ENGLAND

the introduction of competition; and minimization of the state through privatization and contracting out of functions.

In their attempts to reform local government, the Conservatives delveddeep into the NPM shopping basket, deploying all four of Pollitt andBouckaert’s reform strategies at various times over the following two decades.The “maintenance” strategy was manifested in tight controls on local government spending. The level of grants paid to local councils by centralgovernment decreased in real terms, and ministers took new powers torestrict the amounts by which local politicians could raise local taxes.

“Minimization” was evident in the way in which local government wasstripped of key functions. Councils lost control over post-16 education andtraining, and responsibility for regeneration programs was transferred to newbodies which were placed under the control of business people and communityrepresentatives who were directly appointed by central government. The 1988Education Reform Act took away councils’ ability to determine what wastaught in their local schools and introduced a national curriculum. Schoolswere encouraged to opt out of local-authority control altogether and receivetheir funding directly from central government. Those which chose not togo down this route nevertheless received an increasing proportion of theirfunding in the form of central government grants that were dedicated to particular initiatives favored by ministers. Council-house tenants were givena legal right to buy their home at discounts of up to 70 per cent of the truemarket value. Local authorities were left with the poorest-quality accom-modation, and their remaining tenants were given the option to transfer theresidual housing stock to registered social landlords, subject to the outcomeof ballots which used a controversial system that was widely seen as beingweighted in favor of a “yes” vote.

Marketization played a key role in the government’s plans to improve theeconomy, efficiency and effectiveness of local services. From 1980 onwards,local authorities were required by law to expose specified services to com-petitive tendering. A council’s own workforce could only be awarded a contract if they submitted a lower bid than private-sector competitors andgenerated a specified return on capital. The number and range of the ser-vices to which this legislation was applied was increased over time, startingwith manual services such as waste collection and housing management but eventually encompassing a wide range of professions. The attempt to expose services to competition proved deeply unpopular with councils, whichregarded it as an attack on their autonomy and their employees’ pay andconditions. Some engaged in elaborate maneuvers designed to abide by the letter of the law whilst making their services unattractive targets for potential bidders. Private contractors meanwhile complained of a lack of “fair competition”.

Alongside these initiatives, the government also sought to “modernize” local government by introducing private-sector management techniques. As

noted above, there had already been moves to encourage a more corporateapproach. Now the government sought to encourage the adoption of strategicplanning and more systematic performance management. The Audit Commis-sion was initially created to oversee the financial auditing of local authoritiesbut was soon charged with specifying national performance indicators. Coun-cils had to report these data on a regular basis, and the Commission usedthem to compile performance league tables.

Networked local governance

The Labour government that came to power in 1997 also had ambitious plansfor reforming public services. However, unlike their immediate predecessors,ministers increased spending on public services. Local authorities receivedlarge real-terms increases in their budgets as part of a strategy which the government called “invest and reform”. In return for additional resources,ministers expected to see far-reaching changes in the ways in which publicservices were designed and delivered. Administrative systems had, they said, to be updated, and management practices must be overhauled. In thepast, services had too often been designed around the convenience of theproviders. Henceforth they would need to focus on the needs of users (CabinetOffice 1999).

In the health service and in education, ministers turned to a combinationof “modernization” and “marketization” to achieve improvements. Modern-ization initiatives included internal restructuring, the introduction of new information and communications technology and new forms of external inspection (Martin and Davis 2008; Walshe 2008). Marketization involvedthe encouragement of competition between service providers. Although theyinitially rejected the Conservatives’ policy of creating an internal market in the National Health Service, Labour ministers later gave patients the optionto choose the hospital at which they were treated. The theory was that, facedwith the need to attract patients, hospitals would behave like businesses andin the process become more responsive to the needs and aspirations of their“customers”. Similarly, the government encouraged choice between schoolsin the belief that this would put pressure on teachers and governing bodiesto “drive up” standards in order to attract students.

In contrast to the policies that were applied to health and schools, com-petition was not a significant feature of New Labour’s approach to reforminglocal government. One of its first acts was to announce the abolition of therequirement for councils to subject their services to competitive tendering.In its place ministers introduced a new duty of “Best Value”, which requiredcouncils to put in place arrangements to secure continuous improvement by reviewing the efficiency and effectiveness of their services in consultationwith users, citizens and other local interest groups (Martin 2001). Enforcedcompetition was seen as having been counterproductive. Whilst it had

“made the costs of services more transparent”, the “detailed prescription ofthe form and timing of competition led to unimaginative tendering, and oftenfrustrated rather than enhanced real competition” (DETR 1998: clause 7.22).Instead, the government now wished to encourage local councils to forge long-term partnerships with private-sector companies in the provision of awide range of frontline services such as schools and waste management and “back office” functions like legal services, asset management, HR andpayrolls (Entwistle and Martin 2005).

There was a residual element of “minimization” in New Labour’s approachto local government reform. Starved of funds to refurbish their remain-ing housing stock, local authorities were encouraged to transfer them to arm’s-length management organizations. As a result, by 2004 councils heldless than half the number of residential properties which they had owned in1979. Ministers also curtailed local-authority control over schools to an evengreater extent than their predecessors. Having initially toyed with the ideaof abolishing LEAs, they eventually decided to retain them but to encour-age schools to opt out of their control. Failing schools in inner-city areaswere closed down and replaced by independent “city academies” which werepartly funded by private sponsors (such as businesses, faith groups and charities) that took over control of the governing body, curriculum, staffingand management. Other secondary schools were encouraged to apply for status as “specialist schools”, which took them out of local-authority controland allowed them access to private funding. By 2005 the majority had madethis transition.

The centerpiece of New Labour’s approach to reforming local governmentwas through a range of policies that became known as the “modernizationagenda”. These were predicated on the assumption that local councils hadan important role in the delivery of public services but needed to undergowhat a local government white paper in 1998 called a process of “radicalmodernisation”. There needed, it was argued, to be “a fundamental shift of culture”. Councils must “break free from old fashioned practices and attitudes”; “paternalism and inwardness” had to be “swept away” (DETR1998).

The principal agent of the modernization process was an unparalleled and apparently unbridled enthusiasm for top-down performance-monitoring.There was a huge increase in the scale, scope and intensity of external inspec-tion of local authorities. According to the government’s own calculations,the total cost of public services inspection rose steeply from £250 million in1997–8 to £550 million by 2002–3 (OPSR 2003). By 2005 the direct costs of inspection of local government amounted to £97 million per annum(ODPM/HM Treasury 2005). The costs of inspecting education more thandoubled from £88 million to £201 million. Spending on inspection of socialservices increased from £6 million to £11 million. The Audit Commission,which had responsibility for overseeing most other local government services,

saw its budget balloon from £111 million to £217 million and its workforcegrow by almost 90 percent as it recruited an army of new inspectors. For thefirst time all local government services were subject to external scrutiny, andinspectors no longer simply checked that services were meeting minimum standards. They were now charged with assessing a service’s prospects forimprovement. In 2002 the inspection framework was strengthened by the introduction of “Comprehensive Performance Assessments” (CPAs),which scored each council’s overall performance and graded them on a five-point scale ranging from “excellent” to “poor” (Downe and Martin 2007).Councils at the bottom end of the resulting performance league table weresubject to direct intervention, which usually involved the replacement of theirsenior managers and the provision of support and advice by a government-funded improvement and development agency (Martin 2002; Yapp andSkelcher 2007). The increase in external inspection of English local govern-ment was, of course, part of a much broader “audit explosion” which hasbeen a feature of public management across many Western democracies (Power 1997). But it was particularly prominent in the UK (Hood et al. 1999).As has been widely documented, the fragmentation of the local state, asso-ciated with the NPM, led to the displacement of traditional hierarchical forms of coordination, which in turn saw policy-makers turn to the “longdistant mechanics of control” offered by public services audit and inspec-tion (Hoggett 1996).

New Labour also became increasingly concerned with improving horizontalcoordination between local agencies. There was a recognition that the publicwas confused about who was responsible for local services and frustrated athaving to deal with a multitude of providers. The solution was not, it wasargued, to reinvent the local welfare state of the immediate postwar period.What was needed was for local authorities and others to work together to offer more “joined up” services. As the prime minister explained, “It isin partnership with others – public agencies, private companies, communitygroups and voluntary organisations – that local government’s future lies”(Blair 1998: 13). Unlike the previous Conservative government’s vision ofminimalist local government the role of which was to oversee the outsourcingof local services, ministers were not seeking to by-pass councils or to reducetheir role. They wanted to redefine it. Unlike the architects of the local welfare of the immediate postwar period, the Blairite vision of local govern-ment did not assume that it was necessary for local councils to provide local services. Henceforth their role should be “to actively steer processes of co-ordination and collective action across public, private and voluntaryboundaries” (Stoker 2004). To succeed they would have to rely on, and seekto influence, the actions of other bodies. This, it was argued, would enablea more “joined up” approach which was better-suited to dealing with the“wicked” issues that cut across the responsibilities of individual service pro-viders. This model, which became known as “networked community governance”,

envisaged local authorities playing a vital “community leadership” role,operating as the hubs of local service delivery networks and orchestratingthe activities of health service providers, the police, training agencies, busi-nesses and charitable organizations.

The idea of community leadership was not new. Most local politicians saw their primary duty as being to represent constituents’ interests, and mostcouncils had a long tradition of speaking up on behalf of their localities toregional bodies, national government and the European Commission. Theywere also used to coordinating celebrations (such as local carnivals or eveninternational sporting or cultural events) and responses to environmental and economic crises (such as natural disasters or the closure of large localemployers). However, the Blair government’s formal recognition of their community leadership marked an important shift, which was manifested inthree key policies – the creation of Local Strategic Partnerships, a require-ment for local authorities to produce Sustainable Community Strategies, andthe introduction of Local Area Agreements.

Local Strategic Partnerships consist of representatives of the private, public, voluntary and community sectors who meet regularly to review theneeds of their area and formulate plans for joint action to address them. The objectives on which they agree are set out in Sustainable CommunityStrategies, which typically articulate a long-term (five-to-ten-year) vision.Statutory responsibility for preparing strategies rests with local authorities,but they are expected to liaise extensively with the other members of LocalStrategic Partnerships. The strategies also set out plans for consulting withand reporting to local communities. Local Area Agreements are agreed betweena Local Strategic Partnership and central government, and are informed by community strategies and national policy priorities. The latest round ofagreements included up to fifty-three targets based on statutory perform-ance indicators defined by central government departments plus a range ofoutcomes which matter locally. In return for signing up to agreements, localagencies are given greater freedom to pool the funding they receive from central government departments. They also receive performance rewardgrants for hitting targets.

This new approach poses formidable challenges for local authorities. Theyare now being held accountable for the delivery of agreed outcomes over whichthey often have little direct control. Local politicians and local authority officers now have to be able to work across organizational boundaries to a degree that was never expected of previous generations. Staff working in the hierarchical organizations of the past derived their authority from their formal positions and professional expertise. The emergence of net-worked community governance calls for officers and local politicians whoare able to network with, and influence the actions of, other agencies(Sullivan and Skelcher 2002). This “boundary-spanning” behavior calls foran understanding of the often very different priorities, cultures and professional

backgrounds of the diverse organizations that comprise Local StrategicPartnerships (Williams 2002), and for strong negotiating skills and powersof persuasion.

This new style of operating has also blurred traditional lines of account-ability. A recent study found that one-third of the senior local authority managers surveyed believed that partnership working had made it more difficultfor the public to hold local service providers in their area to account (Cowellet al. 2009). However, on the positive side, there is evidence that partner-ship working has made local agencies more directly accountable to each other. In the past, local politicians focused primarily on the activities of local councils. Now that local authorities are being held accountable for theachievement of a much wider range of outcomes, they are looking for waysof scrutinizing the performance of other service providers.

Partnership working of the kind engendered by Local Area Agreementsalso marks a change in the way in which central government holds local agencies to account. In the past, relations between central and local govern-ment were dominated by strong vertical links between individual local services and the central government department which oversaw them. LEAs,for example, reported to the Department for Education, which also oversawthe activities of the Office for Standards in Education (OFSTED) – the bodywith responsibility for school inspection. LEAs often had little interactionwith other council services. OFSTED had minimal contact with otherinspectorates. And the Department for Education had a clearly defined briefbeyond which it rarely strayed far. A similar pattern was apparent in social services, housing, planning and a range of other council services. Here,too, local authority departments had strong vertical links with nationalinspectorates and their associated central ministries which bound togethergroups of practitioners and policy-makers who often came from similar professional backgrounds and had a common sense of purpose (Rhodes 1999). Policies designed to encourage “joined up” working at local level havemade it much more difficult for central government and inspectors to work exclusively within these traditional “silos”.

Attempts to coordinate the activities of inspectorates have included joint reviews of social services undertaken by the (then) Social ServicesInspectorate and the Audit Commission, and more recently “joint areareviews”. Reports by OFSTED and the inspectors of social services have also been used by the Audit Commission to compile CPAs, and there havebeen some mergers of inspectorates. In 2008, for example, OFSTED tookover responsibility for the inspection of all children’s services, adding socialservices to its previous responsibilities for the inspection of schools and colleges. In 2009, Comprehensive Area Assessments were introduced in anattempt to evaluate how well local agencies are working together to deliverthe outcomes specified by Local Area Agreements. The aim is to bring togetherthe judgments made by several different inspectorates into a single report.

However, this process is not easy. Different inspectorates collect differentkinds of performance information in different ways, and it is not a straight-forward task for them to bring together the large volume of data that theyhold, let alone to combine them into an overall judgment about how wellagencies are working together.

Central government departments have also experienced problems dealingin a holistic way with local partnerships. Despite the emphasis on “joinedup” government, central ministries continue to be functionally organized. The services which local authorities provide cut across the responsibilities of a total of eight central government departments and more than twentygovernment ministers, and the mechanisms for coordinating activities acrossdepartments are weak. As a result, there were significant difficulties and delaysexperienced in the negotiation of the first Local Public Service Agreements,which were the forerunners of Local Area Agreements. Local authorities often found it difficult to navigate their way through the “maze” of centralgovernment departments and agreements involved. “Cross-cutting targets were particularly difficult, and revealed a lack of joined-up working betweendepartments in central government” (Sullivan and Gillanders 2005: 563). The potential problems were magnified once it was decided to include in Local Area Agreements the services provided by a much wider range of otherlocal agencies. To alleviate this, Government Regional Offices were used asintermediaries, and conducted the negotiations on behalf of central ministries.This solved many of the difficulties but has not entirely eliminated them, as central government departments have continued to regard agreements primarily as a means of meeting their particular departmental objectives ratherthan of achieving the more “joined up” policy-making and service deliverywhich they now require of local service networks.

Conclusion

The emergence of Local Strategic Partnerships, Sustainable CommunityStrategies and Local Area Agreements is, then, a very significant policy develop-ment. They are evidence of networked community governance in action, proof perhaps that the reforms of the NPM type of the 1980s and 1990s havebeen supplanted by a new approach inspired by a different and distinctiveparadigm. In contrast to Conservative policies two decades ago, the aim is not to by-pass local government or to strip away its functions. Local authorities have in some senses been rehabilitated. But not in their formerrole as direct providers of local services. They are now required to worktogether with other local agencies to a degree that would have been almostunimaginable twenty years ago. This has changed the nature of local govern-ance and the relationships between service providers. It also has importantimplications for central–local relations.

However, it is not a simple case of the previous model of local service delivery having been overturned by a new, wholly different approach. In reality English local government retains many of the features of the postwarlocal welfare state and of the disaggregated local state bequeathed by theNPM-inspired reforms of the Thatcher era. These two states cohabit withnew policies that reflect the new(ish) concept of networked community governance. In truth, councils’ service-delivery role never went away. It was diminished but not extinguished by the policies of the 1980s and 1990s.Councils continue to be big business. They still deliver a vast array of services and employ large numbers of staff – councils in England currentlyemploy more than 2 million people, not far off 10 per cent of the total workforce. Nor has central government suddenly let go of the old levers of control. The new language of central–local “partnership” is important but it falls well short of the kind of autonomy enjoyed by subnational governments in many other European countries. Local Area Agreements haveincentivized a new style of working among local public service networks, butlocal authorities and their partners continue to operate largely as deliveryagents for national government, tied into its priorities through an elaboratesystem of centrally driven performance targets linked directly to their futurefunding.

ReferencesBlair, T. (1998) Leading the Way: A New Vision for Local Government, London: IPPR.Cabinet Office (1999) Modernising Government, London: The Stationery Office.Cowell, R. J., Ashworth, R., Downe, J., Skelcher, C., Bovaird, A. G. and Chen, A.

(2009) The State of Local Democracy: The Impact of Policy Changes on Account-ability and Public Confidence, London: Department for Communities and LocalGovernment.

Department of Environment, Transport and the Regions (1998) Modern Local Govern-ment: In Touch with the People, Cmnd 4014, London: The Stationery Office.

Downe, J. and Martin, S. J. (2007) “Regulation Inside Government: Processes and Impacts of Inspection of Local Public Services”, Policy and Politics, 35 (2):215–32.

Entwistle, T. and Martin, S. J. (2005) “From Competitive Tendering to Col-laboration in Public Service Delivery: A New Agenda for Research”, PublicAdministration, 83 (1): 233–42.

Hoggett, P. (1996) “New Modes of Control in the Public Service”, PublicAdministration, 74: 9–32.

Hood, C. (1991) “A Public Management for All Seasons”, Public Administration, 69 (1): 3–19.

Hood, C., Scott, C., James, O., Jones, G. and Travers, T. (1999) Regulation InsideGovernment, Oxford: Oxford University Press.

Martin, S. J. (2001) “Implementing Best Value: Local Public Services in Transition”,Public Administration, 78 (1): 209–27.

Martin, S. J. (2002) “The Modernisation of UK Local Government: Markets,Managers, Monitors and Mixed Fortunes”, Public Management Review, 4 (3):291–307.

Martin, S. J. and Davis, H. (2008) “The Rise of Public Services Inspection”, in H. Davis and S. J. Martin (eds) Public Services Inspection in the UK, London: Jessica Kingsley.

ODPM/Her Majesty’s Treasury (2005) Securing Better Outcomes: Developing a NewPerformance Management Framework, London: Her Majesty’s Treasury.

OPSR (2003) Inspecting for Improvement, London: Cabinet Office.Pollitt, C. (1995) “Justification by Works or by Faith? Evaluating the New Public

Management”, Evaluation, 1 (2): 133–54.Pollitt, C. (2007) “New Labour’s Re-disorganization: Hyper-modernism and the Costs

of Reform – a Cautionary Tale”, Public Management Review, 9 (4): 529–43.Pollitt, C. and Bouckaert, G. (2000) Public Management Reform: A Comparative

Analysis, Oxford: Oxford University Press.Power, M. (1997) The Audit Society: Rituals of Verification, Oxford: Oxford Univer-

sity Press.Rhodes, R. A. W. (1999) Control and Power in Central–Local Government Relations,

2nd edn, Aldershot: Ashgate.Stoker (2004) Transforming Local Governance: From Thatcherism to New Labour,

Basingstoke: Palgrave Macmillan.Sullivan, H. and Gillanders, G. (2005) “Stretched to the Limit? The Impact of Local

Public Service Agreements on Service Improvement and Central–local relations”,Local Government Studies, 31 (5): 355–74.

Sullivan, H. and Skelcher, C. (2002) Working across Boundaries: Collaboration in PublicServices, Basingstoke: Palgrave Macmillan.

Walshe, K. (2008) “Regulation and Inspection of Health Services”, in H. Davis andS. J. Martin (eds) Public Services Inspection in the UK, London: Jessica Kingsley.

Williams, P. (2002) “The Competent Boundary Spanner”, Public Administration, 80 (1): 103–24.

Wilson, D. and Game, C. (2006) Local Government in the United Kingdom,Basingstoke: Palgrave Macmillan.

Yapp, C. and Skelcher, C. (2007) “Improvement Boards: Building Capability for PublicService Improvement through Peer Support”, Public Money and Management, 27 (4): 285–92.

Part V

GOVERNANCE OF POLICY NETWORKS

POLICY NETWORKS: THEORY AND PRACTICE1

Tobias Jung

While the notion of policy networks can be traced back to Greek philo-sophy (Parry 1969; Kimber and Richardson 1974), it has its modern rootsin the writings of Bentley: at the beginning of the twentieth century he describedgovernment as “networks of activity” (Bentley 1908: 261). At the time, otherauthors toyed with similar ideas. For example, in his Impasse of Democracy,Griffith (1939) advocated the concept of “whirlpools of activity” within thepolitical system. However, it was only during the 1950s that the idea of policy networks started to gain momentum when structural and socio-political changes following World War II led to an increasing complexity inthe organization of government and the governing of society. The key featuresof this included: a move toward a pluralist and collaborative approach todeveloping and implementing policies, division of labor, sectoralization, andfunctional differentiation (Kenis and Schneider 1991; Pappi and Henning 1998).These changes were picked up in the works of authors such as Freeman (1955), Truman (1951), Maass (1951), Dahl (1956), Schattschneider (1935) andLindblom (1965), who, driven by a dissatisfaction with more traditional con-ceptions of the policy process, started to explore the idea of placing groupsof actors with shared interests at the heart of policy-making. Out of this, thepolicy network approach, with its focus on the links between actors involvedin both policy formulation and implementation, developed.

There have been lengthy and not always fruitful debates about the pre-cise role played by the policy networks concept. While some authors appearto have argued that the concept amounts to nothing more than some sortof metaphor (Dowding 1995), others maintain that policy networks exist notonly as a model but also as a real influence on the formulation of policy(Potters and Sloof 1996), acting as “links” amongst the actors in the policyfield (John 1998; Lovseth 2000). Nonetheless, the policy network idea hasbecome increasingly well established and accepted as an analytical tool(Heany 2001a; Albrechts and Lievois 2004). Because of its portable nature,the idea of policy networks can be applied to various areas of public inter-est (John 1998), making it an ideal framework for any policy researcher’s

and practitioner’s toolbox; from their humble origins at the beginning of thetwentieth century, policy networks have emerged as prominent themes acrosscountries, sectors and disciplines (Robinson 2006). Even those who were strongcritics in the past now argue that exploring the networks involved in the policy process is important in analyzing both policy formulation and policyimplementation (Dowding 2000, 2001).

This chapter sketches the terminological pot-pourri surrounding policy networks and provides an outline of the key concepts of iron triangles, issues networks, policy communities and advocacy coalitions; it highlightsthe concepts’ defining characteristics, their advantages and disadvantages.

Terminological confusion

Despite the widespread acceptance and prominent use of the policy networkidea within academic discourse, there is a lack of agreement on networks’defining characteristics, the appropriate terminology, and the concept’s correct application. Assumptions about the ontological, epistemological andmethodological standing of policy networks are diverse: they range from positivistic traditions at one end, to realist and interpretative ones at the other(Marsh and Smith 2001). Those who take a positivist view try to understandnetworks through measures of cohesion, centrality and structure (see, for example, Knoke 1994; Dowding 1995; Milward and Provan 1998; Dowding2001), while proponents rooted in interpretative approaches are interestedin the processes and contents of interactions that take place within networks(see, for example, Marsh and Rhodes 1992; Marsh and Smith 2000; Bevirand Rhodes 2003).

In general, the literature on policy networks appears to have become preoccupied with definitional disputes, conceptual ambiguities and a pro-liferation of typologies (Wolman 1992; Lovseth 2000). As a result, authorshave argued that a “Babylonian variety” of terms and applications has emerged within the field (Börzel 1998), a “terminological jungle in which anynewcomer may plant a tree” (Barnes 1972). Alongside the aforementioned “policy whirlpools” (Griffith 1939), authors have referred to “subsystems”(Freeman 1955), “subgovernments” (Cater 1964), “triangular trading patterns”(Lowi 1962), “sloppy large hexagons” (Jones 1982), “issue niches” (Browne1990), or “epistemic communities” (Haas 1992). The danger of such mush-rooming terminology is that the explanatory power of the concept can easily be inflated (Marin and Mayntz 1991), especially given that there is nocoherent school of thought, glossary of terms, or application. A large num-ber of authors have only a vague idea of what constitutes a policy networkand often fail to state their nebulous assumptions or ideas (McCool 1990;Marsh 1998). It is thus not unheard of that two different authors, while usingthe same term, mean two completely different things or, vice versa, that twoauthors using different terms do actually refer to the same thing (Börzel 1998).

For example, Peterson (2003) highlights incongruities about “advocacycoalitions” between the writings of Sabatier and Jenkins-Smith (1993) andKeck and Sikkink (1998), as well as the different conceptions of policy com-munities in the writings of Rhodes (1990) and Wright (1988).

The aforementioned issues are aggravated by the fact that different countries have traditionally conceived of and used the policy-networkapproach in different ways. There exist at least three different internationalschools of thought relating to policy networks: the American, the British andthe European, the last of which can again be subdivided into Dutch, Frenchand German traditions. These different perspectives can be roughly clusteredunder two headings: the interest intermediation school and the governanceschool (Börzel 1998; Marsh 1998; Marsh and Smith 2000).

The governance school perceives policy networks as an alternative formof governance that by way of its nonhierarchical coordination constitutesan alternative to markets and hierarchies (Börzel 1997, 1998). The problemwith this approach is that it is idealistic, prescriptive and theoretical. Theinterest intermediation school, on the other hand, focuses on analyzing the role policy networks, perceived as various kinds of relationships betweeninterested groups and the state, play in the formulation and implementationof policies. While this perception can be criticized as being overly static, it offers a practical, straightforward way of effectively identifying and out-lining policy networks’ characteristics (Börzel 1997, 1998; Thompson and Pforr2005). To this end, it offers various network models. Despite the termino-logical disparity and variety that can be identified within the literature, theseexisting perspectives are essentially variations on four key models: iron triangles, issue networks, policy communities and advocacy coalitions (see,for example, Hanks 2000).

Iron triangles

Iron triangles’ roots are difficult to identify (Freeman and Stevens 1987), especially as they have confusingly also been referred to as “networks”, “subgovernments”, “subsystems” and “whirlpools” within the literature(Browne and Paik 1993). However, it appears that the concept was origin-ally based on research into agricultural, water and public works policies (Heclo 1978). While the actual components that are nowadays thought of as constituting an iron triangle – interest groups, committees and an executive agency – appear to have been first identified by Cater (1964) in hisexamination of policy-making in Washington’s “subgovernments”, the con-cept as such has become mainly associated with the work of Lowi (1962,1969). Lowi had initially identified “triangular trading patterns” within thepolicy arena (Lowi 1962) and started to refer to the metaphorical conceptof “iron triangles” only in subsequent work, where he expanded on his original idea of “triangular trading patterns” (Lowi 1969).

POLICY NETWORKS: THEORY AND PRACTICE

Traditionally, it has been assumed that iron triangles are very stable tripartitearrangements which operate over a long period of time with little or no out-side interference; in this respect they can be considered to be almost autono-mous as regards their decision-making abilities (Thurber 1991). Iron trianglesnormally control a very narrow and small niche within the policy field, andthe relationships amongst the small number of participants are assumed tobe slow-changing and “mutually advantageous” (Heclo 1978; McCool 1990).Although each member of the triangle receives some sort of benefit from the arrangement, benefits are not necessarily distributed equally (Nachamiasand Rosenbloom 1980). Because of this benefit focus, iron triangles most commonly appear in the case of distributive policies, with each side of thetriangle supporting and complementing the other two (Jordan 1981).

The idea of iron triangles has been considered to be analytically useful inso far as it helps to simplify complex interactional arrangements. As a con-sequence, it has been widely referred to. However, the real-world applicabilityof the concept has been questioned. First of all, it seems to be difficult to identify clearly the three key groups within iron triangles (Browne andPaik 1993). Second, the iron-like structure does not allow for a permeationof it from outside, nor does it account for any relation to the surrounding environment; rather, the iron triangle appears to be some form of secludedinsular arrangement detached from everything else (Stein and Bickes 1995).However, the major criticism of the iron triangle concept has come from Heclo,who has argued that the concept of iron triangles is “not so much wrong as it is disastrously incomplete” (Heclo 1978: 88). Iron triangles cannot takeinto account the huge level of complexity of the policy process: researchersare trying hard to identify the three major players while missing the interactions that take place within the broader networks of people who increasingly have an impact on government (Heclo 1978).

Use of the iron triangle metaphor – even as a “straw man” – oversimplifies political relationships so badly as to hide the very consequential local influences in Congress and its network.

(Browne and Paik 1993: 1075)

Although there have been attempts to make the concept of “iron triangles”more inclusive by adding extra players, illustrated by Cerny’s (2001) movefrom “iron triangles” to “golden pentangles”, one problem with such modi-fications is that the defining nature of the concept is lost and the differencefrom some of the other concepts becomes less clear.

Issue networks

The concept of issue networks, which has also become known under a variety of terms (Heany 2001a), was proposed by Heclo (1978) as a way of

addressing the shortcomings he saw with iron triangles. Although the ori-ginality and conceptual contribution of this concept has been questioned inso far as all it appears to do is to change the label of an earlier concept,“whirlpools” (Freeman and Stevens 1987), it is issue networks that have become widely known. While Heclo considers issue networks to be “almostthe reverse image in each respect” of iron triangles (Heclo 1978: 102), otherauthors have questioned this and argued that issue networks are “not dis-creetly different arrangements from iron triangles” (Jordan 1981: 103).Instead it has been claimed that all they account for are iron triangles withan increased number of participants, larger disaggregation of power, less predictable participants, and both lower cohesion and lower homogeneity(Jordan 1981).

The defining trait of an issue network is the large number of participantsinvolved, ranging from individuals to huge interest-groups. Issue networksoperate at a multitude of levels, and within issue networks the number ofparticipants is in a permanent flux. As a result, no one is really in controlof the issue agenda. Whereas involvement in iron triangles is mainly drivenby materialistic reasons, it is assumed that the underlying reasons for participation in issue networks are emotional or intellectual factors, with members sharing a specific interest. Although on first impression this seemsto imply that issue networks amount to little more than political movements,this is not the case: policy goals are more specific in issue networks than inpolitical movements; and, while one defining trait of political movements isuncertainty about authority within the movement, in issue networks there is no tendency for anybody to obtain perceived or legitimate authority as to what represents the public will (Heclo 1978; Salisbury 1984).

Although the major advantage of issue networks appears to have been thatthe concept offered an alternative to iron triangles (McCool 1990), it hasalso drawn attention to important aspects of the policy process. First of all,it has highlighted that policy is made in communities, the structure, natureand stability of which all influence the policy process. Second, it has drawnattention to the permanent flux within such communities because of playersmoving in as well as out; such changes lead to modifications in agendas, inter-ests and linkages between various issues. Third, the concept brings to lighta more decentralized understanding of power, with no one person being fullyin charge (Heany 2001a).

Nonetheless, there are certain problems with issue networks as an analytical concept. It is incredibly difficult to identify a specific issue network since at any point in time only parts of the network might be active. Consequently, there is no clearly identifiable set of participants, links between participants might fade or be strengthened, and there are noclear-cut boundaries between governmental institutions and their environ-ment. As Heclo (1978) himself recognized when trying to examine issue networks, it is virtually impossible to state where a network leaves off and

its environment starts. Issue networks are therefore “more like amorphousclouds than geometric designs” (Browne and Paik 1993: 1055). As such, theydo not lend themselves to academic studies.

Although the idea of issue networks has been frequently quoted andaccepted (McCool 1990; Heany 2001a), only structures that display some of the features that characterize an issue network appear to have beenidentified within the literature (Sabatier and Jenkins-Smith 1993).

Policy communities

Initially, the idea of policy communities appears to have been a defence of pluralism against its critics and corporatist theory (Jordan 1981), and can be seen as a British reproduction and duplication of the American concepts of iron triangles and issue networks (Jordan 1990). Although it has been pointed out that at the time there was no direct application ofAmerican ideas to the British context, but that the approach taken was simply a description of structures within British policy-making with a sub-sequent recognition of American precedents (Richardson and Jordan 1979),others simply consider it to be a development of the US literature forBritain (Dowding 1995).

Leaving any such disputes aside, this combination of British and Americanschools of thought has had its own merits. While, on the one hand, policycommunities follow the idea of issue networks by widening the number ofparticipants, they also assume stable relationships that mirror the longstandingrelationships within iron triangles, perceived as assisting in the negotia-tion process. Policy communities are thus a special case of stable networks(Jordan 1990); and, were one to draw a continuum of policy network concepts, policy communities would be placed somewhere between iron triangles and issue networks (Hanks 2000).

Within the literature, policy communities are understood to be a sort of “common culture and understanding” within specific policy domains as regards problems and decision-making processes (Dowding 1995: 138).Some of these communities are “diverse and fragmented” while others are“extremely closed and tightly knit” (Kingdon 1995: 118). These commu-nities are made up of specialists in any given policy area (Kingdon 1995); and, if a new policy focus develops, sooner or later a new policy commu-nity will evolve around it (Jordan 1990). As a result, “policy communitiesare swept by intellectual fads” (Kingdon 1995: 127), so that the attentiongiven to certain issues will fluctuate over time (Kingdon 1995): as the focuson one policy area becomes intellectually fatigued and routinized, other areas become more interesting.

One of the main difficulties with the idea of policy communities is the variety of ways in which it has been used. This makes it almost impossibleto come up with a coherent picture (Campbell et al. 1989; Anderson 1990;

Rhodes 1990; Marsh and Rhodes 1992). For example, Grant et al. (1988)identify differentiation, specialization and interaction as characteristics of policy communities, whereas Rhodes (1988) in his examination distinguishesbetween interests, membership, resources, and horizontal and vertical inter-dependence. So, while in its original form the concept was “parsimoniousand thought-provoking”, it has been “complicated and diluted” in subsequentwork (Grant 1995: 34).

More recently, the policy community approach has been dominated bythe work of Rhodes (1988, 1990; Rhodes and Marsh 1992), who considerspolicy communities as only one sort of network amongst professional net-works, intergovernmental networks, producer networks and issue networks.According to this perspective, a policy community’s defining characteristicsare the more restricted number of participants, high-quality interaction ofall groups on all matters related to policy issues, and a dominance of economicor professional interests, with both values and membership persisting overa long period of time (Rhodes and Marsh 1992). However, these character-istics appear to be problematic. At a basic level, it is questionable whetherhigh-quality interaction of all groups on all matters related to policy issuesis even theoretically feasible, and also whether it is possible to maintain theidea of stability over time (Richardson 2000). More important, however, the assumption that only people with some form of expertise can participatein policy communities does not seem to hold, and a focus on mainly economicor professional interests is dubious, with the arising picture resembling an“elite cartel” rather than a policy community (Grant 1995): they appear asexclusionary, hidden gatherings of vested interests that use public resourcesfor private aims (Miller and Demir 2006).

Advocacy coalitions

In order to replace the traditional triangular approach to policy networksand the conventional idea of a stagist policy-cycle, according to which policies can be understood as a set of interdependent phases through whichpolicies go over time, Sabatier – first with Pelkey (1987) and later with Jenkins-Smith (1993) – developed the idea of advocacy coalitions. These form partof a broader framework: the advocacy coalitions framework.

According to the advocacy coalitions framework, networks of policy-makers, within this approach called policy subsystems, are made up of severalcompeting coalitions – normally between two and four (Cairney 1997). Eachof these coalitions consists of a variety of players, ranging from various governmental to private organizations, who share certain core beliefs. Thesecore beliefs are of a fundamental normative and ontological nature, and there-fore not very susceptible to change. Advocacy coalitions will try to translatethese core beliefs, which act like “glue” to hold the advocacy coalitiontogether (Sabatier and Jenkins-Smith 1993), into policies. In order to do so,

they will apply various strategies, such as litigation, lobbying of elected officials,commissioning research or other ways of influencing opinions (Elliott andSchlaepfer 2001). However, opinions on these secondary, more pragmatic and instrumental aspects of how to achieve the policy goal may differbetween various players within the same advocacy coalition and are open to change over time (Sabatier and Jenkins-Smith 1993).

As well as those players within the coalitions, the existence of another groupof actors, the “policy-brokers”, is assumed. These brokers are considered tobe people, such as high civil servants, whose concern it is to keep the levelof political conflict within acceptable limits so that a “reasonable” solutionto a perceived problem is reached (Sabatier and Jenkins-Smith 1993). It appears that it will usually be difficult to distinguish between these brokersand “advocates” since they are based on a continuum.

Outside the policy subsystem there are two factors that provide bothresources and restraints for the actors within. First of all, there are relativelystable parameters that hardly change in the short run, such as the basic distribution of the natural resources, basic attributes of the “problem” area,basic legal structure and the fundamental socio-cultural values and structures.Second, there are those aspects that are open to major changes in the shortrun. These include changes in socio-economic conditions, public opinion, governing coalitions as well as policy decisions, and impacts from other subsystems. It is these aspects which provide the main driving forces behindchanges in policies (Sabatier and Jenkins-Smith 1993: 20–3).

Despite the fact that the advocacy coalitions framework comes with a set of “testable” hypotheses which have been tried through several case studies, the major criticism of this concept has been that it is mainly basedon the American system. It is consequently difficult to apply to any polit-ical system that does not correspond to American-style pluralism (Parsons1995: 200–3). Although an attempt at a “European” version was publishedin 1998 (Sabatier 1998), its usefulness is open for debate. Sabatier essenti-ally appeared to ask European researchers to do their own work: “The real task of European researchers is to develop falsifiable hypotheses basedupon the ACF” (Sabatier 1998: 121). While several studies have tried to applythe concept to European structures, mainly the EU itself (see, for example,Radaelli 1999; Warleigh 2000; Elliott and Schlaepfer 2001; Weber andChristophersen 2002), questions have arisen about the framework’s applic-ability to a European context. For example, Warleigh (2000: 237) concludesthat “Sabatier’s model of the advocacy coalition . . . appears unable entirelyto encapsulate the entrepreneurial dynamics of EU decision-making”.However, a proper critical analysis of the extent to which the advocacy coalitions framework’s original criteria can be extended to the EU is still wanting: to date, most applications of the advocacy coalitions frameworkwithin EU contexts have not so much focused on theory-testing as appliedthe framework as a heuristic device (Fueg 2009).

Leaving aside issues about the advocacy coalitions framework’s trans-ferability, more general problems with this concept arise out of the notionof belief systems. First of all, the grouping of people according to beliefs rather than to importance or influence is problematic. Second, the idea of core beliefs might not explain the reasons why groups form coalitions.Third, core beliefs might not shape the day-to-day operations and actionsof coalitions. Fourth, the distinction between various levels of beliefs is awkward, and it will be virtually impossible to identify a coalition’s core beliefs(Cairney 1997). Fifth, there are questions about whether a priori core beliefsare a valid assumption: people’s beliefs and values are vague, contradictoryand unstable (Hajer 1995). Sixth, since no historical context is developed aspart of the advocacy coalitions framework which allows for an analysis ofcoalition formation (Sabatier and Jenkins-Smith 1993), the development of a coalition’s core beliefs cannot be analyzed (Watt 1997). Given this lack of historical context, the advocacy coalitions framework lends itself more to established policy networks (Sabatier and Jenkins-Smith 1993), andthus does not really take into account how different topics can come on tothe policy agenda; the advocacy coalitions framework might explain policystability better than policy change (Fischer 2003).

Other prominent criticisms outlined by Parsons (1995) have addressed thedistinction between “events” and “stable parameters”: the question is howfar the constraints and resources set for the policy subsystem by these twofactors exist in a subsystem’s physical environment or if they are cognitiveconstructs within the individuals and organizations that make up the sub-system. In addition, there is an assumption that non-elites, such as membersof the general public, have “neither the expertise, nor the time, nor the inclination to be active participants in a policy subsystem” (Sabatier 1999:202). The final concern about the extent to which the advocacy coalitionsframework can contribute to a better understanding of the policy process is its methodological and epistemological limitations: it is firmly rooted in a positivist worldview (Fischer 2003).

Concluding comments

Since the 1950s, policy networks have emerged as a prominent analytical toolwithin academic discourses for exploring policy processes. Unfortunately, a lot of energy has been wasted on debates as to which conceptualization of policy networks is the most appropriate. What is often ignored is the com-plementary nature of the underlying models and the real-world practical application they offer. For example, LaPorte (1996) has drawn attention to the possibilities of approaching networks from different perspectives: from within the network, from above the network, and from the side of thenetwork. The first is an organization-centric view that considers the net-work through the eyes of a key network participant looking out, up, across

and down at other actors with whom it must deal. The second perspectivetakes an elevated position and explores a sector’s contexts and structures fromabove. The final approach looks at the network from the side, as a playerwho has, as of yet, limited stakes in the network under observation but whooperates from a similar institutional status level to some of those within that network. Similarly, by taking an organization-centric view, Mikkelsen(2006) has highlighted how using policy networks as a strategic, analyticaltool might offer network players a better understanding of the context inwhich they operate, and their options for pursuing and achieving theirstrategic objectives. In both instances, the models of iron triangles, issue networks, policy communities and advocacy coalitions can assist. While noneof the concepts is perfect, and it is unlikely that any concept will ever be,each highlights different facets of the same phenomenon (Hudson et al. 2007):they represent a continuum and should be considered as complimentary. Incombination, these four concepts draw attention to aspects that range fromrigid, longstanding and limited structures to open and amorphous filigrees;they address different degrees of ease of entry into a network and the levelof structuration therein. As such, they are helpful in identifying and teasingout key facets of a policy network, independently as to whether one’s inter-est in a network is of an academic or a practical nature.

Note1 This chapter is revised and developed from one presented at the UK–Japanese

Comparative Workshop on Local Governance, held in Tokyo in December 2008,with the support of the Daiwa Foundation, the GB-Sasakawa Foundation and theJapan Foundation Endowment Committee.

References

Albrechts, L. and Lievois, G. (2004) “The Flemish Diamond: Urban Network in theMaking?”, European Planning Studies, 12 (3): 351–70.

Anderson, J. E. (1990) Public Policymaking: An Introduction, Boston, Mass.: HoughtonMifflin.

Barnes, J. (1972) “Social Networks”, Addison-Wesley Module in Anthropology, 26: 1–29.Bentley, A. F. (1908) The Process of Government, Chicago, Ill.: University of Chicago

Press.Bevir, M. and Rhodes, R. A. W. (2003) Interpreting British Governance, London:

Routledge.Börzel, T. A. (1997) “What’s So Special about Policy Networks? An Exploration of

the Concept and Its Usefulness in Studying European Governance”, EuropeanIntegration online Papers (EIoP), 1 (16), http://eiop.or.at/eiop/texte/1997–016a.htm

Börzel, T. A. (1998) “Organizing Babylon: On the Different Conceptions of PolicyNetworks”, Public Administration, 76: 253–73.

Browne, W. P. (1990) “Organized Interests and Their Issue Niches: A Search forPluralism in a Policy Domain”, Journal of Politics, 52: 482–526.

Browne, W. P. and Paik, W. K. (1993) “Beyond the Domain: Recasting NetworkPolitics in the Postreform Congress”, American Journal of Political Science, 37 (4):1054–78.

Cairney, P. (1997) “Policy Studies: Theory and Practice”, http://www.pas.ac.uk/CPS/1997/cair.pdf [accessed 2 June 2004].

Campbell, J. C., Baskin, M. A., Baumgartner, F. R. and Halpern, N. P. (1989)“Afterword on Policy Communities: A Framework for Conceptual Research”,Governance: An International Journal of Policy, Administration, and Institutions, 2: 86–94.

Cater, D. (1964) Power in Washington: A Critical Look at Today’s Struggle in theNation’s Capital, New York: Random House.

Cerny, P. (2001) “From ‘Iron Triangles’ to ‘Golden Pentangles’? Globalizing the PolicyProcess”, Global Governance, 7: 397–410.

Dahl, R. A. (1956) A Preface to Democratic Theory, Chicago, Ill.: University of ChicagoPress.

Dowding, K. (1995) “Model or Metaphor? A Critical Review of the NetworkApproach”, Political Studies, 43: 136–58.

Dowding, K. (2000) “How Not to Use Evolutionary Theory in Politics: A Critiqueof Peter John”, British Journal of Politics and International Relations, 2 (1): 72–80.

Dowding, K. (2001) “There Must Be End to Confusion: Policy Networks, Intel-lectual Fatigue, and the Need for Political Science Methods Courses in BritishUniversities”, Political Studies, 49: 89–105.

Elliott, C. and Schlaepfer, R. (2001) “The Advocacy Coalition Framework: Applica-tion to the Policy Process for the Development of Forest Certification in Sweden”,Journal of European Public Policy, 8 (4): 642–61.

Fischer, F. (2003) Reframing Public Policy: Discursive Politics and DeliberativePractices, Oxford: Oxford University Press.

Freeman, J. L. (1955) The Political Process: Executive Bureau–Legislative CommitteeRelations, New York: Doubleday.

Freeman, J. L. and Stevens, J. P. (1987) “A Theoretical and Conceptual Reexamina-tion of Subsystem Politics”, Public Policy and Administration, 2 (1): 9–24.

Fueg, O. C. (2009) “The Advocacy Coalition Framework Goes to Europe. AnAmerican Theory and Its Application across the Pond”, paper presented at theAnnual Convention of the Political Studies Association, Manchester, 7–9 April.

Grant, W. (1995) Pressure Groups, Politics and Democracy in Britain, London:Harvester Wheatsheaf.

Grant, W., Paterson, W. and Whitson, C. (1988) Government and the Chemical Industry,Oxford: Clarendon Press.

Griffith, E. S. (1939) The Impasse of Democracy, New York: Harrison-Hilton.Haas, P. M. (1992) “Introduction: Epistemic Communities and International Policy

Coordination”, International Organization, 46 (1): 1–35.Hajer, M. A. (1995) The Politics of Environmental Discourse: Ecological Modern-

ization and the Policy Process, Oxford: Oxford University Press.Hanks, C. D. (2000) “Reexamining the Subsystem Concept”, Political Science,

Texas A&M University: 185.Heany, M. T. (2001) “The Evolution of Interest-group Networks: A Study of the

US National Health Policy Domain, 1977–2001”, Department of Political Scienceand Harris School of Public Policy, Chicago, Ill.: University of Chicago.

Heclo, H. (1978) “Issue Networks and the Executive Establishment”, in A. King (ed.) The New American Political System, Washington, DC: American EnterpriseInstitute.

Hudson, J., Lowe, S., Oscroft, N. and Snell, C. (2007) “Activating Policy Networks:A Case Study of Local Environmental Policy-making in the United Kingdom”,Policy Studies, 28 (1): 55–70.

John, P. (1998) Analysing Public Policy, London: Pinter.Jones, C. O. (1982) The United States Congress: People, Place, and Policy, Homewood,

Ill.: The Dorsey Press.Jordan, A. G. (1981) “Iron Triangles, Woolly Corporatism, and Elastic Nets:

Images of the Policy Process”, Journal of Public Policy, 1: 95–123.Jordan, A. G. (1990) “Sub-governments, Policy Communities and Networks: Refilling

the Old Bottles?”, Journal of Theoretical Politics, 2: 319–38.Keck, M. E. and Sikkink, K. (1998) Activists beyond Borders: Advocacy Networks

in International Politics, London: Cornell University Press.Kenis, P. and Schneider, V. (1991) “Policy Networks and Policy Analysis: Scrutiniz-

ing a New Analytical Toolbox”, In B. Marin and R. Mayntz (eds) Policy Networks:Empirical Evidence and Theoretical Considerations, Frankfurt am Main: CampusVerlag.

Kimber, R. and Richardson, J. J. (eds) (1974) Pressure Groups in Britain: A Reader,London: Dent.

Kingdon, J. W. (1995) Agendas, Alternatives, and Public Policies, Boston, Mass.: Little,Brown.

Knoke, D. (1994) Political Networks: The Structural Perspective, Cambridge:Cambridge University Press.

LaPorte, T. R. (1996) “Shifting Vantage and Conceptual Puzzles in UnderstandingPublic Organization Networks”, Journal of Public Administration Research andTheory, 6 (1): 49–76.

Lindblom, C. E. (1965) The Intelligence of Democracy, New York: The Free Press.

Lovseth, T. (2000) “Informal Networks – What Matters? On Policy Networks andthe New Institutionalism in Organization Theory”, paper prepared for EuropeanConsortium for Political Research Joint Sessions, Copenhagen.

Lowi, T. J. (1962) Legislative Politics USA, Boston, Mass.: Little, Brown.Lowi, T. J. (1969) The End of Liberalism: Ideology, Policy, and the Crisis of Public

Authority, New York: W. W. Norton.Maass, A. (1951) Muddy Waters: The Army Engineers and the Nation’s Rivers,

Cambridge, Mass.: Harvard University Press.McCool, D. (1990) “Subgovernments as Determinants of Political Viability”, Political

Science Quarterly, 105 (2): 269–93.Marin, B. and Mayntz, R. (eds) (1991) Policy Networks: Empirical Evidence and

Theoretical Considerations, Frankfurt am Main: Campus Verlag.Marsh, D. (1998) “The Utility and Future of Policy Network Analysis”, in

D. Marsh (ed.) Comparing Policy Networks, Buckingham: Open University Press.Marsh, D. and Rhodes, R. A. W. (eds) (1992) Policy Networks in British Govern-

ment, Oxford: Clarendon Press.Marsh, D. and Smith, M. (2000) “Understanding Policy Networks: Towards a

Dialectical Approach”, Political Studies, 48 (4): 4–21.

Marsh, D. and M. J. Smith (2001) “There Is More than One Way to Do PoliticalScience: On Different Ways to Study Policy Networks”, Political Studies, 49: 528–41.

Mikkelsen, M. (2006) “Policy Network Analysis as a Strategic Tool for theVoluntary Sector”, Policy Studies, 27 (1): 17–26.

Miller, H. T. and Demir, T. (2006) “Policy Communities”, in F. Fischer, G. Millerand M. S. Sidney (eds) Handbook of Public Policy Analysis, Boca Raton, Fla: CRCPress.

Milward, H. Brinton and Provan, K. G. (1998) “Measuring Network Structure”, PublicAdministration, 76 (Summer): 387–407.

Nachamias, D. and Rosenbloom, D. H. (1980) Bureaucratic Government USA, NewYork: St Martin’s Press.

Pappi, F. U. and Henning, C. H. C. A. (1998) “Policy Networks: More than aMetaphor?”, Journal of Theoretical Politics, 10 (4): 553–75.

Parry, G. (1969) Political Elites, London: George Allen & Unwin.Parsons, W. (1995) Public Policy: An Introduction to the Theory and Practice of Policy

Analysis, Cheltenham: Edward Elgar.Peterson, J. (2003): “Policy Networks”, Political Science Series, no. 90, Vienna:

Institute for Advanced Studies.Potters, J. and Sloof, R. (1996) “Interest Groups: A Survey of Empirical Models That

Try to Assess Their Influence”, European Journal of Political Economy, 12: 403–42.Radaelli, C. M. (1999) “Harmful Tax Competition in the EU: Policy Narratives and

Advocacy Coalitions”, Journal of Common Market Studies, 37 (4): 661–82.Rhodes, R. A. W. (1988) Beyond Westminster and Whitehall, London: Unwin Hyman.Rhodes, R. A. W. (1990) “Policy Networks: A British Perspective”, Journal of

Theoretical Politics, 2 (3): 293–317.Rhodes, R. A. W. and Marsh, D. (1992) “New Directions in the Study of Policy

Networks”, European Journal of Political Research, 21 (1): 181–205.Richardson, J. (2000) “Government, Interest Groups and Policy Change”, Political

Studies, 48: 1006–25.Richardson, J. and Jordan, G. (1979) Governing under Pressure: The Policy Process

in a Post-parliamentary Democracy, Oxford: Martin Robertson.Robinson, S. E. (2006) “A Decade of Treating Networks Seriously”, Policy Studies,

34 (4): 589–98.Sabatier, P. A. (1998) “The Advocacy Coalition Framework: Revisions and Rele-

vance for Europe”, Journal of European Public Policy, 5 (1): 98–130.Sabatier, P. A. (ed.) (1999) Theories of the Policy Process: Theoretical Lenses on Public

Policy, Boulder, Colo.: Westview Press.Sabatier, P. A. and Jenkins-Smith, H. C. (1993) Policy Change and Learning: An

Advocacy Coalition Approach, Boulder, Colo.: Westview Press.Sabatier, P. and Pelkey, N. (1987) “Incorporating Multiple Actors and Guidance

Instruments into Models of Regulatory Policymaking: An Advocacy CoalitionFramework”, Administration and Society, 19 (2): 236–63.

Salisbury, R. (1984) “Interest Representation: The Dominance of Institutions”,American Political Science Review, 78 (1): 64–76.

Schattschneider, E. E. (1935) Politics, Pressures and the Tariff, New York: Arno Press.Stein, R. M. and Bickes, K. N. (1995) Perpetuating the Pork Barrel: Policy Sub-

systems and American Democracy, Cambridge: Cambridge University Press.

Thompson, G. and Pforr, C. (2005) Policy Networks and Good Governance: A Discus-sion, Perth, Australia: Curtin University of Technology, School of Management.

Thurber, J. A. (1991) “Dynamics of Policy Subsystems in American Politics”, in A. Cigler and B. A. Loomis (eds) Interest Group Politics, Washington, DC: Con-gressional Quarterly Press.

Truman, D. B. (1951) The Governmental Process: Political Interests and PublicOpinion, New York: Knopf.

Warleigh, A. (2000) “The Hustle: Citizenship Practice, NGOs and ‘Policy Coali-tions’ in the European Union: The Cases of Auto Oil, Drinking Water and UnitPricing”, Journal of European Public Policy, 7 (2): 229–43.

Watt, R. N. (1997) “Policy Networks in Theoretical Perspective”, http://www.pas.ac.uk/CPS/1997/watt.pdf [accessed 2 June 2004].

Weber, N. and Christophersen, T. (2002) “The Influence of Non-governmentalOrganisations on the Creation of Naturea 2000 during the European PolicyProcess”, Forest Policy and Economics, 4: 1–12.

Wolman, H. (1992) “Understanding Cross National Policy Transfers: The Case of Britain and the US”, Governance: An International Journal of Policy, Adminis-tration, and Institutions, 5 (1): 27–45.

Wright, M. (1988) “Policy Community, Policy Network and Comparative IndustrialPolicies”, Political Studies, 36 (2): 593–612.

POLICY NETWORKS IN PRACTICE: THE DEBATE ON

THE FUTURE OF AMSTERDAMAIRPORT SCHIPHOL

Menno Huys and Joop Koppenjan

1. Introduction

The continuously growing airport Amsterdam Airport Schiphol is situatedin a densely populated region near the economic heart of the Netherlands.Since the 1950s the Dutch government has struggled with the trade-offbetween the economic importance of Schiphol and the environmental impactof the increasing air traffic. In 1988 policy-making had reached a dead-lock. Actors participating in the policy debate could not agree on the futureof the airport, nor on the kind of policies that were needed to regulate environmental effects. In an attempt to break the deadlock, government formulated the so-called dual objective: the ambitious growth strategy of the airport would be combined with the simultaneous realization of environ-mental objectives. For the next twenty years (1988–2008), the policy discussionrevolved around the translation of this dual objective into concrete policymeasures.

In this contribution we analyze the policy-making process regardingSchiphol in the period 1988–2008 by using a policy network perceptive. The Schiphol case is characterized by a large set of public and private stake-holders with diverging and conflicting interests that are engaged in a highlypoliticized power game. The network approach provides us with a con-ceptual framework to describe and analyze the strategic interactions withina setting of relatively autonomous but interdependent actors, and to explainthe course and outcomes of these interactions. Furthermore, using the policynetwork theory in the case of Schiphol seems even more appropriate, giventhe attempt of government to combine growth and improvement of the environment. It is network theory that suggests managing conflicts by transforming zero-sum games into win–win situations. Also, new interactivepolicy approaches that have been applied during the past twenty years can

be viewed as forms of network governance, by which government involves a wider set of actors in the policy-making process. Here network theory presents itself as a governance mode, rather than as theory of governance.The normative implications of network theory are brought in practice as network governance: a specific policy or management style to be applied inmulti-actor settings.

In short, the Schiphol case provides us with an opportunity to assess boththe analytical and practical value of the network approach for complex policy-making issues.

Section 2 briefly outlines the central conceptual, theoretical, normative andprescriptive notions of the policy network theory as we apply them in ouranalysis of the policy-making process regarding the future of Schiphol. Sub-sequently, section 3 presents an analysis of the process and the outcome ofthe interactions among actors involved in policy-making regarding Schiphol,using concepts of the network approach as a heuristic tool. Section 4 providesan additional analysis, demonstrating the kind of explanations, evaluationsand lessons that the policy network perspective provides. Section 5 wrapsthings up.

2. Applying the policy network approach to the Schiphol casus

The policy network approach provides a theoretical perspective for analyz-ing, evaluating and improving interaction process regarding complex issueswithin networks of mutually dependent actors. It differs from other policyapproaches like rational choice or New Public Management in that it includesstrategic and institutional factors in the analysis of policy-making and stressesnot so much effectiveness or efficiency as important success criteria, but collaborative advantages, legitimacy and trust (Pierre 2000; Teisman 2000;Huxham 2000; Fukuyama 1995).

Analyzing policy-making in networks: the network approach as heuristic tool

Policy processes are conceptualized as political interaction processes or policy games, in which interdependent actors with diverging or conflictinginterests, perceptions and strategies try to influence the policy process andits outcomes (Axelrod 1984; Scharpf 1997). These interactions take place inone or more arenas: places where actors meet, using strategies to influencepolicy-making (Allison 1971; Crozier and Friedberg 1980). In doing so, suchactors may form coalitions either to support or to oppose certain policies(Sabatier 1988).

As far as these activities are embedded in earlier interactions, actors arepart of a policy network and share outlooks, rules, language and trust that

help them to interact and that reduce transaction costs (Rhodes 1997;Williamson 1998). However, if representatives of different policy networksmeet in specific policy games, institutional features are not compatible andadd to the complexity of the interaction process (Koppenjan and Klijn 2004).

The policy process does not evolve in a linear way, going through a seriesof logically structured stages (Parsons 1995). Rather it is a policy gamethat can be analyzed by distinguishing a number of rounds of interactions.These rounds evolve in an erratic, unstructured way (Teisman 2000). In eachround, actors explore problems and solutions, and look for opportunities toreach a win–win outcome (Dery 1984). However, this is far from simple: diverg-ing perceptions and conflicts of interest may result in a deadlock (Olson 1965).Then, again, breakthroughs may occur, which give the game new impulsesand, to a large extent, determine the conditions for the next round in thepolicy game.

Interaction processes are considered successful if actors succeed in discoveringcommon interests and arrive at win–win situations. In addition to substantiveoutcomes, interaction may produce institutional effects: improved or deteri-orated relationships and levels of trust between actors, with repercussionsfor future interactions (Ostrom 1990).

Explaining policy-making in networks

The network perspective seeks to explain why parties succeed or fail in real-izing collaboration and win–win outcomes. Four explanations are especiallysuggested:

1. Social causes: The ability or inability of actors to coordinate their go-alone strategies. This may be influenced by the quality, intensity orlack of interaction between interdependent actors, and the presence orabsence of social variety. Deadlocks, for instance, may be broken by theentrance of new actors, or the formation of new coalitions (Termeer andKoppenjan 1997).

2. Cognitive causes: The ability of actors to discover collaborative advan-tages or win–win situations, e.g. by inventing new solutions, reframingthe problem or changing the scope for policy-making (Rein and Schön1992). This ability may be influenced by the lack or presence of cognitivevariety: new ideas, knowledge or information (Van Eeten 1999).

3. Institutional causes: The lack or presence of institutions that reduce therisks and cost of interaction such as rules, shared convictions, norms andvalues, a shared language and trust.

4. Management causes: The absence or presence of adequate network governance. It is to be distinguished from Command and Controlapproaches, aimed at imposing one-sided solutions. It also differs fromNew Public Management strategies aimed at reducing interdependencies

POLICY NETWORKS IN PRACTICE

by unbundling and contracting. Network governance consists of effortsto address the interdependencies of actors by involving them in the policy-making process and by arranging and managing this process. It includesthe fulfillment of roles of process architect, facilitator, broker and medi-ator (see Susskind and Cruikshank 1987; O’Toole 1988; Klijn 2008).

Evaluating policy-making and lessons drawn: the policy network theory as normative approach

The policy network approach may be considered an empirical theory, aimedat describing and explaining interactions in complex network settings. As such,the theory facilitates the analysis of policy processes from a strategic per-spective, paying attention to the role of perceptions, strategies and conflict,the institutional factors underlying these, and the outcomes of these powergames in terms of allocation of values over various stakeholders, of whichsome may win and others may lose. Normative assumptions of the networktheory include the consideration of win–win outcomes of network processesas success. Recommendations are aimed at realizing these: promoting inter-action and collaboration, and suggesting supportive institutional arrangementand network governance to enhance these (Kickert et al. 1997; Agranoff and McGuire 2003; Keast et al. 2002; Soerenson and Torfing 2007). It is important to distinguish between the policy network approach as an empiricaltheory and as a prescriptive model applied in practice. As an empirical theory,the policy network approach may have a wider applicability and validity than as prescriptive model. It is quite possible to analyze the implications of hierarchical interventions or new public management strategies from a policy network perspective. Moreover, network governance strategies usedin practice may well be at odds with the theoretical and normative implica-tions of the policy network theory. So the network policy approach as empirical and normative theory may well be used to assess network govern-ance practices.

Applying policy network theory to the Schiphol casus

Before describing and analyzing twenty years of debate about the future of Schiphol, it is important to underline the difference between network theory as empirical and as prescriptive model. Given the involvement of various interdependent actors with diverging and conflicting interests, the policy network theory seems to be an appropriate framework for analyzingthe Schiphol case. As far as the application of network theory in practice is concerned, the case study displays a particular Dutch take on network governance, which is rooted in the Dutch culture of consensualism. This culture of consensus-building can be related to the specific situation of the

Netherlands. Its low lands, high water-levels and dense population called forcollaboration between people in order to provide enough suitable land forbuilding (Edelenbos 2000). Moreover, the relatively large outside world forcedthe small Netherlands to cooperate internally, strengthening the consensusorientation (Dijkink 1990). Over the years the lack of space has made it moreand more important to attune the many different claims for infrastructure,housing, industry, nature, recreation, agriculture to one another, giving riseto several institutions for collaborative policy-making and planning (Faludiand Valk 1994). The Dutch consensualism is characterized by a high degreeof corporatism, wherein a few powerful interest groups (i.e. labor unions,large multi-nationals) are included in the national policy-making processes(Van Waarden 1999). Finally, the Dutch consensualism is also highly pragmaticin nature, resulting in commonly used dispensations, policy experiments, policy evaluations and tolerance of illegal drugs and prostitution (Van Wijk2007). It is against the background of this specific consensus-oriented cul-ture, with hints of corporatism and pragmatism, that the particular Dutchtake on network practice is to be understood.

However, this does not imply that the study has no value for other casesembedded in another context. First of all, the Schiphol case is analyzed usingpolicy network theory as empirical and normative theory. This results in an analysis that, given the generic, international theoretical foundations ofthis theory, may have specific findings, but is not confined in its approachor implications to the particularities of the Dutch context. Second, althoughnetwork governance matches the Dutch institutional context, which may explain the popularity of both the empirical and prescriptive variants of policynetwork theory among Dutch academics and practitioners, the recognitionof the need of network governance and the application of these types of man-agement strategies are not restricted to the Netherlands (Richardson 1982;Rhodes 1990; O’Toole 1988; Keast et al. 2002). As a result, the findings regard-ing the application of network governance in this case may hold lessons thatare relevant for other settings in other countries. Third, it can be argued thatthe Schiphol case is a typical example of what might be labeled a critical case.In essence, a critical case can enhance the “generalizability” of case studies.Such cases often hold more information than randomly selected representa-tive cases, because they activate more actors and more basic mechanisms inthe situation studied. The critical case can be defined as having strategic import-ance in relation to the general problem. Such cases offer information, whichpermits logical deductions of the type “if this is (not) valid for this case, thenit applies to all (no) cases” (Flyvbjerg 2001: 79). Of course, it is difficult toidentify a critical case beforehand. There are no universal methodologicalprinciples available by which one can identify a critical case. The best thingto do is to look for a case which is likely either to confirm or falsify pro-positions and hypotheses (Flyvbjerg 2001; Ragin 1992).

The conflictory nature of the issues discussed in the Schiphol case and thelack of trust make it such a critical case. Several authors have commentedthat the Schiphol policy network is characterized by low levels of trust (cf. Van Boxtel and Huys 2005; Bröer 2006; Van Eeten 1999; Van Gils et al.2009; RMNO 2009; Tan 2001). Trust is regarded as perceptions of the goodintentions of other actors. It concerns an expectation about the intention of another actor, and that intention concerns the expectation that the otheractor will respect the interests of the “trusting” actor (Koppenjan and Klijn2004: 83; Nooteboom 2002). When expectations are repeatedly violated, the trust in another actor must be reviewed or may even disappear entirely.

36 L3 L

HaarlemHaarlemHaarlem

HoofddcrpHoofddorp

18 R18 R18 R

18 C18 C18 C

36 C36 C36 C

060606

an090909

242424

18 L18 L18 LBuitenvelderbaanBuitenvelderbaanBuitenvelderbaan 272727

222222

Kaagbaan

36 R6 R36 R

040404

ZwanenburgZwanenburgZwanenburg

AmsterdamAmsterdamAmsterdam

AmstelveenAmstelveenAmstelveen

AalsmeerAalsmeerAalsmeer

Figure 21.1 Spatial situation Schiphol region 2006.Source: Milieu– en Natuur Planbureau, 2006.

A high degree of trust within networks decreases transaction costs betweencooperating actors, enhances the probability that actors will exchange information, and encourages learning and innovation. In the case of wickedproblems, where there is need for intensive coordination, and where levelsof uncertainty are high, trust becomes an essential factor in the creation ofsuccessful governance networks. Without trust, it may be “just networks”that we are left with, holding little joint decision-making despite intensiveinteraction patterns (cf. Keast et al. 2005; Parker 2007). In essence, trade-offs between different spatial claims (i.e. airport infrastructure including noise protection zones and other spatial claims like housing, industrial sites, landside accessibility and green areas) need to be made in the denselypopulated Schiphol region owing to the lack of space (see Figure 21.1), butthe high level of distrust makes it very complicated to do so in an effectiveand legitimate way. Given these specific characteristics of the Schiphol case, it is all the more relevant to examine the application of network govern-ance strategies. If a specific type of network governance works for theSchiphol case, it is likely to work in less conflicted and fixated policy situ-ations elsewhere.

In the next section we analyze the policy debate of Schiphol as a series of three rounds, using the central concepts of the policy network theory. Insection 4 the success or failure of the process is established by determiningto what extent win–win solutions have been created. In addition, explanationsare sought by referring to cognitive, social, institutional and managementcauses. Next, lessons are drawn on how to improve the quality and outcomeof policy debate. In addition, the case study may provide us with clues asto what extent the normative assumptions of this theory hold in practice.

3. Twenty years of debate on the future of Schiphol: three rounds of policy-making

In 1988 the Dutch government established a new policy strategy forSchiphol, the so-called mainport strategy (cf. Van Duinen 2004; Pestman 2001).Schiphol was perceived to be an important cornerstone of further economicdevelopment of the Netherlands, while at the same time it was stated thatthis development should not proceed at the expense of the environment(Ministry of Physical Planning 1988: 185). In 1989 the quest to translate thedual objective of the mainport strategy into concrete policy measures couldbegin. In this section we describe the Schiphol policy debate, distinguishingthree rounds of policy-making:

1. The preparation of the Spatial Planning Key Decision (PKB) (from 1988till 1997);

2. The drafting of the new Schiphol law (from 1998 until 2006);3. The revision of the Schiphol law (2006 until 2008).

3.1. Round 1: The PKB (Spatial Planning Key Decision) process: 1988–97

This policy round revolved around the operationalization and implementa-tion of the dual objective in a so-called Spatial Planning Key Decision(Planologische Kernbeslissing, PKB). The PKB procedure is an extensive decision-making procedure that results in a legally binding national spatialplanning decision (Van Buuren et al. 1999). Government chose to use theROM-method, a new participative policy approach designed to develop integral tailored plans for specific areas that could count on wide public support (Ministry of Physical Planning 1988). It consisted of two steps: develop-ing a start covenant and translating this covenant into concrete measures.After these preparatory steps, the formal policy-making could begin.

The composition of the policy arena

During the development of the start covenant (the so-called Plan of ApproachSchiphol and Surroundings, Plan van Aanpak Schiphol en Omgeving, PASO),actors from three different networks – i.e. the aviation network, the spatialand infrastructural planning network, and the local /environmental network– converged in the newly established policy arena (see Figure 21.2). TheDepartment of the Environment (DGM) of the Ministry of VROM (Housing,Spatial Planning and the Environment) was put in charge of setting up thenew arena, replacing the Ministry of Transport’s Governmental AviationAgency (Rijksluchtvaartdienst, RLD) that was hitherto in charge of theSchiphol affair (Tan 2001). From March 1989 onwards, DGM approachedthe main stakeholders one by one, resulting in participation of the Provinceof North Holland, the Municipality of Haarlemmermeer and the SchipholAirport Authority (Driessen 1995). Only after this initial network was formeddid DGM invite the RLD. DGM’s strategy first to establish some goodwillamongst the other parties worked out well. This way there was some pres-sure for RLD to join in. These five actors established the so-called steeringgroup, which started to prepare the start covenant.

Step 1: negotiating the start covenant

During the negotiations about the content of the covenant, Schiphol man-aged to make its own masterplan, with a detailed investment plan for thecoming fifteen years, part of PASO. In order to do so, Schiphol had insistedon involving the Department of Economic Affairs of the municipality ofAmsterdam in the Steering Group (Werther 1992). On 21 September 1989the now seven parties of the Steering Group signed the Start Covenant. Thedual objective was very much defined in terms of mainport development versus noise pollution: the airport was allowed to grow, as long as the noise

pollution would be diminished. Some actors that were not involved in theprocess, like local residents, the environmental parties (most notably theStichting Natuur en Milieu, SNM), the Planning Department of the muni-cipality of Amsterdam and RARO (the independent advisory council of theMinistry of VROM), raised concerns about the feasibility of the dual object-ive and the rather economic foundation of the perspective (PAU 2000).

Step 2: negotiating the PASO report

In the remainder of PASO, KLM (Royal Dutch Airlines, KoninklijkeLuchtvaartmaatschappij) was included in the Steering Group, and the pro-ject group was extended with some airlines and municipalities (Driessen 1995).Very soon it became clear that it was impossible to reduce noise pollutionwith the assumed traffic growth of 6 percent a year. Therefore the dual objec-tive was further reformulated. Earlier it was stated that Schiphol developmentshould fit within the existing environmental standards. Now the more generalnotion was adopted that the quality of the living environment had to beimproved. Furthermore, the criterion for noise pollution was adapted. It wasno longer derived from the amount of Ke in the housing areas around Schiphol,but from the amount of houses within the 35 Ke zone.2 Research had shownthat this criterion offered more opportunities to combine the desired growthof aviation (of 6 percent per year) with an improvement of the noise situ-ation (Ket 2000).

The Steering Group ignited negotiations about the amount of houses thatwas desirable within the 35 Ke zone. The environmental coalition, consist-ing of DGM, North Holland and Haarlemmermeer, assumed that there wouldbe 40 million passengers in 2015, and proposed a maximum of 9,000 houseswithin the zone. The pro-growth or economic coalition, consisting of RLD,Schiphol, KLM and Economic Affairs, opposed this since it would frustratefurther growth to 50–60 million pax. in 2015. They wanted at least 11,500houses within the contours in 2015. The impasse was solved by distinguish-ing between a short- and a long-term perspective (Tan 2001). For the shortterm, the Steering Group agreed that 15,000 houses would be allowed withinthe 35 Ke. For the long term, after 2003, when the new fifth runway wouldhave come into operation, parties agreed with the amount of 10,000 houses(Ket 2000; PAU 2000).

Still, the noise issue was not totally resolved. The discussion about nightflights remained at an impasse. The pro-environment coalition argued thatnight flights should be ended. The pro-growth coalition stressed that such aradical measure would seriously harm the position of Schiphol vis-à-vis otherEuropean airports. As the sense of urgency to develop the final version ofPASO increased, the actors decided to postpone the issue of night flights (Tan2001). However, the province of North Holland tried to bring the issue backon to the agenda. Moreover, it called for establishing maximum transport

volumes that would serve as hard limits to growth. Eventually, the Ministryof Spatial Planning (DGM) proposed to evaluate the night regime and thegrowth of flights in 1993. The Provincial Board hesitantly agreed with theseterms, since it feared to be excluded from future rounds of Schiphol policy-making. In April 1991 the PASO covenant was signed (PAU 2000; BureauPAU 2000). The final PASO contained 111 measures (Projectbureau Main-port and Milieu Schiphol (PMMS) 1991; VINO 1988; Werther 1992). Mostimportant, it was agreed that a fifth runway was necessary. This runway wouldconsiderably reduce the amount of houses within the 35 Ke zone (from 16,500to 10,000).

The formal decision-making: translating PASO proposals into the Spatial Key Decision (PKB)

Next, formal decision-making could finally begin, meant to translate PASOinitiatives into a legally binding national Spatial Key Decision (PlanologischeKern Beslissing, PKB). The Steering Committee remained in place, with themain difference that the secretary-general of the Ministry of Transportationtook over the management role of DGM. The most important issues debatedduring the PKB procedure concerned the definition of a mainport and theregulation of noise pollution.

The critical mainport barrier referred to the minimum amount of trafficthat was needed to sustain mainport operations. Different scenarios were developed. The Steering Group decided on 6 April 1993 that the lowest critical mainport barrier would be used as the point of departure for the remainder of the PKB process. The growth levels were much lower than those made by Boeing, KLM or in Schiphol’s Masterplan, but this scenariooffered the best possibilities for reconciling the environmental (i.e. noise pollution) and mainport objectives.

As regards the discussion on noise nuisance, a distinction was made betweenthe short-term four-runway system and the mid-term five-runway system. Forthe short-term four-runway system, the environmental objective was solelydefined in terms of noise pollution: if there were fewer than 15,000 houseswithin the 35 Ke zone, the environmental objective was achieved. For thefive-runway system (after 2003) the noise criterion was adapted (i.e. the levelof noise pollution should be improved in comparison with 1990) and addi-tional limits for external safety, air pollution and stench would come intooperation (PKB, pt 4, 1995: 8). The noise limits were calculated for 235 pointsthat made up the contour. For each point, the amount of decibels had to belower than 35 Ke during daytime, and no more than 15,100 houses wereallowed within the 35 Ke zone. During night-time (2300–0600), only 10,100houses were allowed within the area exposed to more than 26 decibels.

However, it proved impossible to fit the desired mainport development within the noise limits, especially during daytime (15,000 houses within 35 Ke

zone). The solution for the noise problem popped up from a rather unex-pected side. A new calculation model for noise was developed. This new modelregistered the actual flight routes of the airplanes by their radar tracks. Theold model calculated these flight tracks. By a lucky coincidence, this resultedin different noise contours wherein only 14,900 houses fell into the zone (Ket 2000). As regards the five-runway system, it was argued that all of thenew criteria would be met (PKB, pt 4, 1995). The environmental parties, especially SNM, repeatedly questioned the noise norms used and the waythe contours were calculated, but with no effect whatsoever. Another envir-onmental party, Milieudefensie (Environmental Defence), adopted a moreradical strategy. Besides organizing several protests, they bought a strategicpiece of cropland in April 1994, just where the fifth runway was to be located.From November 1994 onwards, they started to plant trees there and createdthe Bulderbos (Bulder Forest). This forest was to prevent the possibility forexpropriation (De Kruijf 2002).3

The outcome of the PKB process and its aftermath

At the end of 1995 the PKB was politically ratified. The Lower House wasnot totally convinced about the rather low growth that was assumed in PKB. In the end it was decided that capacity limits would be introduced toprevent unfettered growth. Passenger numbers were not allowed to exceed44 million in any year, and cargo tonnes were not allowed to exceed 3.3 mil-lion in any year. In the PKB of 1995 it was agreed that the noise limits wouldbecome effective from 1997 onwards, to give the sector parties sufficient timeto adapt their daily operations.

Schiphol immediately exceeded the limits of several enforcement points in 1997 and 1998. In an evaluation report of the PASO/PKB process it wasconcluded that the forecasts were unrealistic (far too low), and probably usedbecause they made the dual objective possible (Algemene Rekenkamer 1998see also CPB 1998). The local residents and the environmental parties didnot tolerate the excesses and started the first of many juridical proceduresthat would heavily frustrate further decision-making. In Figure 21.3 the structure of the policy arena is presented.

Still, the amount of houses within the daytime zone (12,800) was far below the maximum (15,000). This discrepancy was to be attributed to thespecific shape of the calculated and legally binding noise zone. The In’t Veld Committee, which was to assess the effectiveness of the current noise system (TK 25466, Nr 9), indicated that it was rather inefficient that the 35 Ke zone was exceeded at a few points, especially points where nobodylived (the so-called pastureland), whereas at most other points a lot of spacewas left. If these “absurdities” were to be repaired, noise pollution could bereduced and capacity could be increased. In practice, this implied a yearlyincrease of 20,000 flights, starting in 1997 (360,000), and resulting in 460,000

The TOPS process: impasses regarding extra flights and noise regulation

TOPS’ first assignment was to advise on making the yearly increase of 20,000 flights possible without increasing noise pollution. Schiphol took theinitiative and designed a new noise zone in close cooperation with other sector parties. A research report of a consultancy firm concluded that theenvironmental situation would deteriorate considerably. Subsequently the environmental parties rejected the new zone, and the TOPS actors failed toreach an overall agreement (Weggeman 2003).

Initially, TOPS would advise on a new noise regulative system for the five-runway system (2003 onwards). Instead the ONL (Onderzoek nationaleLuchthaven, Research Program National Airport) established by the RLD(Ministry of Transport) in March 1999 took over the initiative. In order tomake sure that the new system could be put into force by 2003, a first designhad to be ready at the end of 1999. In September 1999, ONL presented itsalternatives for a new system, which were rejected by the sector. The sector, especially Schiphol, did not want to discuss the issue within TOPS(Weggeman 2003). As RLD and Schiphol negotiated the new regulative system, TOPS was sidelined.

At first, ignoring TOPS was not likely to be an efficient strategy, con-sidering that Milieudefensie owned pieces of land where the new runway was to be constructed. However, the Cabinet had created a new emergencylaw, Noodwet Procedures Vijfde Baan (Emergency Act Procedures FifthRunway), in January 1999 (Ministry of Transportation 2002). As such,Schiphol was legally empowered to acquire the missing pieces of land fromMilieudefensie. Furthermore, the Cabinet opted for designing an entirely new Schiphol law, instead of a new PKB decision. This allowed for a lesscomprehensive procedure than the PKB (TK 27603, Nr 6/2000, p. 7).

The preparation of the new law

In the Cabinet’s report Future of the National Airport (Toekomst van deNationale Luchthaven, TNL 1999) the new regulative system was presented.It is stated that the new system offers equal protection to the PKB systemthat it comes to replace (TNL 1999: 6). In the new system, still no enforce-ment points were located in the outer areas. The Lower House had someserious doubts about the level of equivalence between the old and the newsystems. In a response, the Cabinet installed a new independent committeeof noise experts, chaired by Professor Berkhout (Commissie DeskundigenVliegtuiggeluid, CDV) (Staatscourant, 20 juni 2000).

However, the CDV was very critical of the new noise system. As a con-sequence, the CDV was sidelined. The Minister of Transportation arguedthat the committee meddled in affairs for which others were responsible

(Berkhout 2003). However, the Environmental Impact Assessment Committee(EIA committee) also expressed its doubts about the value of the new noise system. The environmental actors and the local residents respondedby developing an alternative Schiphol Act, the Citizen Initiative Act(Burgerinitiatief wet, 2001). Since the Lower House held some serious doubts,the minister agreed to add fifty new enforcement points within the 35–20 Kezone (the outer areas) as a matter of compromise (ONL, 2001, Nr 9). The Upper House ratified the new Act in July 2002 (cf. EK 27603), butdemanded that the new system be evaluated within three years (2006) (motionBaarda, 2002). Schiphol and KLM were satisfied with the new Act since it allowed them to put the new fifth runway (Polderbaan) needed further to facilitate hub-development into operation in February 2003.

The outcome of this round of policy-making

In February 2003 the Polderbaan was opened, and the Schiphol Act came intoeffect. The new regulative system did not work out very well, though. Theamount of complaints about noise annoyance increased. Schiphol, KLM andATM complained that it hampered mainport development, because the rigidflight rules did not allow for flexible runway use, so the additional capacitycould not be optimally used (TK 29665, Nr 13/2005). The Committee RegionalDiscussion Schiphol (Commissie Regionaal Overleg Schiphol, CROS)4 thatwas established in 2003 to negotiate about the development of concrete noise-reducing measures indicated that measures for reducing noise hindrance couldnot be implemented within the regulative system. In essence, just as with theold PKB system, the new system was not based on the actual flight routes,resulting in the exceeding of limits in some enforcement points. The rigidlaw made it impossible to adjust the system to the actual flight patterns, whichcould result in more capacity and less noise (see Figure 21.4).

In its evaluation report on the new Schiphol Act, the Cabinet concludedthat the criterion of equality of protection in the old and the new regulatorysystem was met, but that improvements should be made regarding the useof the airport (i.e. facilitate more flights) and the reduction of noise nuisance,especially in the outer areas (Evaluation Report, 2006). Figure 21.5 showsthe structure of the Schiphol policy arena anno 2006.

3.3. Round 3: revising the Schiphol law (2006–8)

At the beginning of 2006, the Cabinet announced that they would frame a covenant in close consultation with the aviation sector, in which firm and maintainable agreements would be made for reducing noise hindrance,while increasing capacity (Cabinet 2006: 5). As a response to this announce-ment, the sector (i.e. Schiphol Group, KLM), the regional authorities(Haarlemmermeer, Amsterdam and Province of North Holland) and the

The composition of the arena in this round

In 2006 central government installed the so-called Alders table, named afterits chairman Hans Alders. Actors gathering around this negotiation tablewere the sector parties (Schiphol Group, KLM/AF, ATM), the regional andlocal authorities (North Holland, Amsterdam, Haarlemmermeer, but also the municipalities of Amstelveen, Uitgeest), the CROS, and the Ministriesof Transport and of Spatial Planning. The CROS insisted on two places forlocal residents, which added an innovative element to the forms of networkgovernance used so far: for the first time, local residents were included inthe formal negotiations about the future of Schiphol. The environmental parties did not take part in the negotiations. After TOPS they revived theirold strategy of protesting, lobbying and taking legal action. This time, theCabinet did not run formal policy trajectories alongside the table. Just as in 1999, the first assignment was to find short-term solutions for revising the existing noise system, of which noise limits had been exceeded in a fewenforcement points in 2006 and 2007 (TK 29665, Nr 43/2007). As in all other years, excesses were tolerated by the Cabinet, relating the breaches to force majeure and in an anticipation of the revision of the rigid system(TK 29665, Nr 43/2007; TK 29665, Nr 68/2007; TK 29665, Nr 103/2008).

The table of Alders in action: preparing the advice for the short term and the long term

Schiphol suggested five alternatives for future development, which would serveas a point of departure for the negotiations (Startnotitie MER 2007).5 Thetwo most prominent alternatives were based on facilitating further growthto 600,000 flights. The local residents formulated a sixth alternative, aimingfor qualitative growth. Only the hub-operations of AF/KLM were allowedto grow to a maximum of 500,000 flights. Freight traffic and charters were to be removed to the regional airports of Lelystad and Eindhoven, noadjustments to the prevailing limits of noise pollution were tolerated, andan improvement of noise protection of residents in the outer areas was calledfor. Schiphol agreed to include the alternative in the EIA, mainly to avoidearly frustration of the process.

In June 2007 the Alders table presented its advice for the short term.6 Amongother things, it was proposed to allow some more pollution at some and less pollution at other enforcement points of the Airport Planning Decree(Luchthavenverkeersbesluit), so 480,000 flights will become possible in 2010(with 435,000 in 2007). The Cabinet was content about the advice since itwas unanimous and since it secured further growth for the short term.

Before the second Alders round began, the collection of platforms of localresidents (Vereniging Gezamenlijke Platforms, VGP) indicated that theywere not adequately represented by the two CROS members (who were thought

to represent all residents). One implication of the short-term advice was that it would result in considerably more noise pollution in some of the outerareas (VGP letter to the Cabinet, 25 June 2007). Therefore, the Cabinet decidedthat the VGP was to be included at the Alders negotiations for the mid-term(TK 29665, Nr 70/2007).

The negotiations about the mid-term revolved around the most desirabledevelopment scenario for 2018/2020, of the six presented in the EIA of 2007, and two updated covenants on noise hindrance reduction and spatialquality for the mid-term. However, since two renowned knowledge institutes,NLR (Airspace Laboratory) and the MNP (Environmental Planning Agency),could not agree upon the validity of the new calculation methods for deter-mining the new limits for the enforcement points, the Alders table was askedto develop a new system.

The outcome of the Alders table: unprecedented unanimity

In October 2008, the Alders table presented its final report about the mid-term development. There were two main conclusions. First, Schiphol wasallowed to grow to 510,000 aircraft movements in 2020, with a maximumof 32,000 flight movements during the night regime. Furthermore, additionalair traffic – approximately 70,000 flight movements – was to be removed to the regional airports of Lelystad and Eindhoven. More specifically, all hub-related traffic would be concentrated on Schiphol, whereas the freightcarriers and the charters were to be removed to the regional airports. Second,the actors agreed that the only way to get rid of the endless technical dis-cussions regarding noise nuisance was to get rid of the enforcement points,including the complicated calculation methods and the hampering main-tenance system (i.e. every year a new reason was found to tolerate the sevenexcesses in ten years’ time). In the new system noise is to be regulated throughthe development of new and flexible rules for runway use, restricted flightpaths and flight heights, designed to ensure the lowest amount of noise pollution, a maximum amount of flight movements (510,000), and limits for the amount of houses and people exposed to serious pollution. In someareas no measures can be developed to reduce noise pollution. By means of compensation, the Cabinet has reserved 10 million euros for improvingthe quality of life in these areas.

In the end, almost all actors were pleased with the agreement. By allow-ing only hub-related traffic, AF/KLM would get sufficient possibilities toexpand in the near future (Van Gils et al. 2009). The regional public author-ities were also very pleased, since the selective development sits comfortablywith their so-called metropolitan strategy, which they developed from 2003onwards (cf. BRS 2006) The local residents were also pleased because addi-tional flights would be removed to other airports. Still, the other part of theagreement (i.e. about the new noise system) initially caused disagreement among

appointments about noise pollution do not allow for any further expansion.The first protests groups have already emerged (e.g. the platform of alarmedresidents of Flevoland, VIVF, around Lelystad airport). The actors aroundthe regional airport think it is rather strange that decisions are made without their involvement. Second, the airlines that are supposed to leaveSchiphol, like EasyJet (which is the second-largest carrier at the airport), donot have any intention to do so. Finally, it can be argued that the expectedtraffic growth (580,000 in 2020) is rather low. A report commissioned by theMinistry of Transport published in 2006 expected around 720,000 flights in 2020. Just as in 1995, it seems that a rather low projection was needed toenable an agreement between the dual objectives.

4. Analyzing process, outcomes and governance strategies

In this section the success or failure of the policy debate concerning the future of Schiphol is discussed from a policy network perspective. This isdone by asking to what extent the actors succeeded in engaging in forms of collaboration, resulting in win–win solutions, more specifically in realiz-ing the dual objective aimed at reconciling growth with preservation of thequality of the environment. In addition, explanations are sought by refer-ring to cognitive, social, institutional and management causes. Finally, the implications of these findings are discussed. Which lessons can be drawn?

4.1 The process and the outcome of the policy game assessed

The three subsequent rounds of policy-making followed more or less similarpatterns. The policy rounds started with attempts by government to involveyet another category of dissatisfied stakeholders in the policy-making pro-cess, using forms of network governance. First, the arena was broadened toinvolve provincial and local governments; next, environmental parties wereinvolved; and eventually residents were invited to join the negotiations. Yet,with the exception of the Alders table, in the processes that followed, actorsdid not succeed in carrying collaboration through and instead ended up in a power game in which the environmental coalition was outmaneuvered.The inability to combine both objectives was concealed by distinguishingbetween short-run measures and long-term solutions; choosing unrealistic scenarios that fitted the perspective of combining growth and environment;adapting new calculating methods which showed that further growth wouldnot harm the environmental performance; making exceptions to agreed-uponregulations; limiting the discussion to noise only, etc.

In contrast with the dual objective, the result was systematically biased:the trade-off between growth and environment always ended up in favor of the first. The failure to reach agreements and to enforce the agreed-uponenvironmental measures repeatedly resulted in impasses and all-time lows

of distrust in both the national government and its policy framework. So, from a policy network perspective, the interaction process and the net-work governance efforts were far from successful, as was the substantive outcome of the process. Recently, however, actors seem to have become more successful in collaboration and reaching mutually agreed upon out-comes, although this was also to be attributed to a “blackmail” strategy (i.e. residents had to sign at the expense of being excluded from the policyarena), external forces (i.e. the economic recession led to lower growth intraffic volumes, making the sector agree to lower growth alternatives), andremoving part of the problem to other airports that were not included in thenegotiations.

4.2 Explaining the policy game and its outcomes

Cognitive cause: limited policy space and lack of substantive variety

An important explanation for the way the policy process evolved lies in thelimited policy space for realizing win–win situations during most of the time.As a result, old assumptions, arguments, research findings and delineationswere continuously reproduced, reducing the room for innovative ideas. Forexample, the main assumptions underlying the growth–noise debate – i.e. thepossibility of the dual objective, the conceptualization of the environmentaleffects in terms of noise, the conceptualization of noise hindrance in termsof decibels, the presentation of air-traffic growth as something inevitable –were never questioned during the twenty years of discussion. During the Alderstable the scope of the debate was widened in a geographical way: shiftingflights to regional airports. The future has to prove whether this scope changewill result in win–win situations, but since the underlying assumptions haveremained intact, still leading to a narrow framing of the problem (growthversus noise), it can be expected that the old feuds will come to the fore onceagain in the near future.

Social causes: power games reproducing impasses and distrust

The policy debate revolved around two rather stable coalitions (one pro-growthand one anti-growth). The introduction of new actors did not change thisdivide. This lack of social variety reinforced the cognitive fixation mentionedabove. Furthermore, despite the opening up of the arena for newcomers, attitudes remained adversarial, resulting in a power game, reproducing andenhancing opportunistic behavior, impasses and distrust. Only during the third round did social variety emerge: new coalitions were formed, and actors used adapted different strategies, e.g. residents replacing protest by apro-active attitude. It seems that both sector and region have become moreaware of their interdependence, resulting in a more collaborative stance.

However, a deeper understanding of the negotiations works to nuance thiscollaborative stance: local residents were put under pressure to support theagreement; the economic recession made Schiphol lower its growth expecta-tions; and the removal of the problems to local airports was only possibleowing to the exclusion of those airports and the airlines involved during thenegotiations.

Institutional causes: networks in the background

In the policy process, representatives of three policy networks met: the inter-national aviation network (Schiphol, KLM, and other airlines and their allies),the spatial development and infrastructure network (the national, regionaland local governments), and the local network (environmental parties and[platforms of ] local residents). Since they did not have a joint institutionalbackground, they had a hard time finding ways to interact constructively.Furthermore, proposals to regulate Schiphol lagged behind the invest-ment decisions on further growth, taken in the aviation network. As such,Schiphol and KLM were always a step ahead of the actors in the regulatoryarena, determining its agenda, rather than being regulated by it. The relationships were further biased by the asymmetrical division of resources(e.g. management skills, capacity, expertise and information) among par-ticipants, the members of the aviation sector having a clear advantage compared to the other parties. The collaboration between Schiphol and theregional actors during the third policy round is an indication of the gradualincorporation of Schiphol Airport and its traditional allies into a larger and extending regulatory network. The consensus over shifting growth onto other regions is exactly what “traditional, closed” policy networks areexpected to do: dividing the gains of collaboration internally and trans-ferring the cost to outsiders not represented within the network (Jordan 1990).For Schiphol and KLM, collaboration meant trading off autonomy againstthe advantages of a negotiated environment.

Management causes

The inconsistent way central government applied network governancemethods contributed to the failed collaboration and unbalanced outcomes.This inconsistency stemmed from the involvement of different (parts of ) ministries, each representing different networks and interests, which succes-sively dominated the management of the process (e.g. the EnvironmentalAgency of the Ministry of Housing, Physical Planning and the Environ-ment, the RLD of the Ministry of Transport and the Ministry of EconomicAffairs). Especially the gap between the promises of central government about regulating noise nuisance and the failure to enforce these generatedfurther distrust. Moreover, the Ministry of Transport was committed to the

growth objective, and was not prepared to refrain from the old practice of taking one-sided decisions, supportive of the sectors’ growth ambitions,thus marginalizing the interactive policy arena. Only during the last perioddid central government operate coherently by waiting for proposals from the Alders table, thus contributing to its success.

4.3 Lessons to be drawn

Building on the above analyses from a policy network perspective, one ofthe lessons to be drawn is that network governance should focus on furtherresolving the cognitive and social fixations in the debate, a process alreadystarted during the Alders table. Introducing variety is a way to achieve this,for instance by furthering reflection on the basic assumptions underlying theSchiphol debate (e.g. by changing the agenda by asking questions like: Isthe dual objective still desirable? Is environment only about noise nuisance?Is growth self-evident?). And by further broadening the arena to includeregional actors in the policy-making process, who will be affected by the shiftof flights to regional airports. The analysis also emphasizes the role of trustas a precondition for collaboration. Building and maintaining trust is a cornerstone of network governance. As argued in section 2, without trust itmay be “just networks” that we are left with, holding little joint decision-making despite intensive interaction patterns (cf. Keast et al. 2005; Parker2007). For government, this means that it should find a way of dealing with its substantive commitment to specific interests, on the one hand, andthe need of impartiality that is related to the role of process manager, onthe other. Prerequisites are not imposing solutions on other stakeholders, not marginalizing interactive arenas, managing expectations and preventingdifferences between stated and revealed preferences (Koppenjan and Klijn2004). Another cause of distrust to be dealt with is the in-transparency oftechnological models and the role of experts, making it hard for non-expertsto judge research results and the impacts of policy measures. Reverting totechnocratic models may be a way of drawing up a smokescreen behind whichunpopular decisions can be made by the national government. Moreover, as we have seen, it might result in a world of calculations that does not match real-world developments. This greatly enhances distrust. The specificframing of the debate (i.e. growth v. noise) does not work to restore relation-ships of trust, either. This framing narrows the solution space and preventsthe development of longstanding agreements, making it almost impossiblefor interactive governance arrangements to succeed.

Finally, the case study also illustrates that engaging in network collab-oration and network governance is a long-term process, which only starts paying off after a while, since all parties involved have to go through a learning curve, by which they only gradually and often reluctantly inter-nalize the new ways of acting and thinking that are required.

5. Conclusion

Applying the policy network approach to the Schiphol case has resulted inan assessment of the policy-making process regarding Schiphol Airport as almost a twenty-year stagnating interaction process, in which parties failed at reconciling growth ambitions with environmental objectives. Onlyrecently have actors seemed to develop some new ideas and ways of doingthat might open up new opportunities for future policy-making. The ana-lysis also provides us with explanations: the asymmetric power-relations, theopportunistic strategies of Schiphol Airport, and the inconsistent way centralgovernment applied network governance methods contributed to the failedcollaboration and unbalanced outcomes. As a result, initiatives to attemptto build trust were frustrated, and this seriously hindered the developmentof successful network practices. The Schiphol case shows that the win–winoutcomes the policy network theory advocates are sometimes hard to realize.The positive turn the process recently took may be explained by changes in strategy on the side of Schiphol and the government, owing to ongoingsocietal pressure, growing awareness of interdependencies and learning beha-vior owing to ongoing interactions. A more skeptical approach would statethat the broadening of the scope to regional airports should be considereda shift of the cost of ongoing growth to new stakeholders. This was madepossible under representation of the interests of these stakeholders in the inter-active arenas of the Alders table. This could be considered a shortcomingof the network governance mode used and a lesson for improvement.

However, the analysis is not conclusive in determining whether this lackof success can be attributed to the inappropriate application of network governance practices or to the inappropriateness of network governance itself.In the first situation, efforts to improve network governance practices are tobe considered. In the latter situation it would be wise to give up further attemptsto reconcile growth and noise nuisance. It may well be that certain situationssimply lack a common ground for the development of such outcomes. Perhapsthis uncovers an important limitation of network governance and policy network theory. The persistency of actors to stick to deliberation and inter-action despite the repeated failure of these governance strategies may indeedsupport the idea of a typical Dutch take on network practice in this case.

On the other hand, merely using win–win situations as evaluation criteriafor interaction in network-like settings may be too rigid. Given the growthof the airport, the standstill regarding the environmental impacts may be considered as an accomplishment, which would not have been realized without the double objective. Seen in this way, the double objective shouldrather be seen as a management tool than as a performance measure. Thisobjective framed the policy debate, increasingly forcing the pro-growthcoalition to pay attention to the environmental impacts of their activities.The willingness of actors to remain involved in the policy-making process,

despite frustrations and disappointment, indicates that they themselves valuethese interaction processes as worthwhile and see them as an opportunityfor safeguarding their interests. What is more, the transformation of the troublesome interaction into a more or less genuine network is reason for amore positive judgment. The prolonged and not always fruitful interactionsresulted at least to a certain extent in durable relationships and common under-standings, offering improved conditions for jointly addressing problems infuture policy-making rounds.

As stated in section 2, we perceive Schiphol as a critical case, a typicalexample of a wicked problem concerning competing values like economy andecology that can be found in many other sectors and many other countriestoday. Its critical character makes us argue that if the network approach asmanagement approach works for Schiphol, then it has potential to work forother (similar) cases, too. Of course, institutional conditions (e.g. a multi-party system, a history of collaboration and neo-corporatism, a culture ofconsensualism and depoliticized, pragmatic conflict-solving) may constrainthe application of network governance strategies to certain settings, but not in such a rigid way that these governance modes should be consideredto be an exclusive Dutch practice.

However, as far as the success of network governance in the Schiphol case is concerned, our analysis is not conclusive. Before making decisive state-ments on the applicability of network governance practices in the Schipholcase and beyond, it would be wise to compare the performance of Schipholin terms of growth and environmental performance with airports in othercountries, regulated by more hierarchical or New Public Management orien-tated governance modes.

Notes1 Acronyms: AEA = Association of European Airlines; EASA = European Aviation

Safety Agency; IATA = International Air Transport Association; ICAO = Inter-national Civil Aviation Organisation; ATAG = Air Transport Action Group; ATM= Air Traffic Management; VNV = Vereniging Nederlandse Vliegeniers (pilots);NS = Nederlandse Spoorwegen (Dutch Railways); ANWB = Algemene Neder-landse Wielrijders Bond; WNF = Wereldnatuurfonds (World Nature Fund);IPCC = Intergovernmental Panel on Climate Change; EU = European Union.

2 Since the 1960s noise pollution is expressed in Ke (Kosten eenheid – cost unit) in the Netherlands, where 35 Ke coincides with approximately 60 decibels.

3 Places with special environmental, monumental or ecological value are moredifficult to expropriate in the Netherlands.

4 The CROS consisted of all municipalities that fell within the 20 Ke zone (althoughrepresented in nine clusters), nine local residents, the Province and the sector parties (CROS 2003).

5 Startnotitie MER “Verder werken aan de toekomst van Schiphol en de regio”, April2007, SG en LVNL.

6 See letter of Hans Alders to the Ministers of Transportation and of SpatialPlanning, 13 June 2007.

References

Agranoff, R. and McGuire. M. (2003) Collaborative Public Management: NewStrategies for Local Governments, Washington, DC: Georgetown University Press.

Algemene Rekenkamer (1998) Groeicijfers Schiphol, Den Haag: Sdu Uitgeverij.Allison, G. T. (1971) The Essence of Decision, Boston, Mass.: Little, Brown.Axelrod, R. (1984) The Evolution of Cooperation, New York: Basic Books.Berkhout, G. (2003) Dossier Schiphol: Relaas van een falend democractisch process,

Value report.Boxtel, M. van and Huys, M. G. (2005) “Unraveling Decision Making about the

Future Development of Amsterdam Airport Schiphol”, paper presented at 45thEuropean Regional Science Association, Amsterdam, 23–7 August.

Bröer, C. (2006) Beleid vormt overlast, hoe beleidsdiscoursen de beleving van geluidbepalen [Policy Shapes Annoyance: How Policy Discourses Shape the Experience ofAircraft Sound ], Amsterdam: Aksant.

BRS (Bestuurlijke Regiegroep Schiphol) (2006) Middellange termijn visie ontwikke-ling Schiphol en Omgeving. Haarlem.

Bureau PAU (2000) Schiphol en de Ruimtelijke Ordening: De ontwikkeling van hetSchipholcomplex 1980–2000, Groningen/Den Haag.

Buuren, P. J. J. van, Backes, C. W. and de Gier, A. J. J. (1999) HoofdlijnenRuimtelijk Bestuursrecht, Deventer: Kluwer.

Centraal Planbureau (1998) Toekomstige Ontwikkeling vervoersontwikkelingen opLuchthaven Schipol: een second opinion. CPB Externe notitie 98/23, The Hague.

Crozier, M. and Friedberg, E. (1980) Actors and Systems: The Politics of CollectiveAction, Chicago, Ill./London: University of Chicago Press.

Dery, D. (1984) Problem Definition in Policy Analysis, Lawrence, Kan.: UniversityPress of Kansas.

Dijkink, G. J. (1990) Beleidenissen, politieke en ambtelijke cultuur in Nederland1965–1990, Groningen: STYX Publications.

Driessen, P. P. J. (1995) “Activating a Policy Network: The Case of Schiphol”, in P. Glasbergen (ed.) Managing Environmental Disputes: Network Management asan Alternative, Dordrecht: Kluwer.

Duinen, L. van (2004) “Planning Imagery”, PhD thesis, PrintPartners Ipskamp.Edelenbos, J. (2000) Proces in vorm; procesbegeleiding van interactieve beleidsvorming

over locale ruimtelijke projecten, Utrecht: Lemma.Eeten, M. van (1999) Dialogues of the Deaf: Defining New Agendas for Environ-

mental Deadlocks, Delft: Eburon.Faludi, A. and van der Valk, A. J. (1994) Rule and Order: Dutch Planning Doctrine

in the Twentieth Century, Dordrecht: Kluwer.Flyvbjerg, B. (2001) Making Social Science Matter. Cambridge University Press.Fukuyama, F. (1995) Trust, the Social Virtues and the Creation of Prosperity, New

York: The Free Press.Gils, M. K. A. van, Huys, M. G. and de Jong, B. (2009) De Nederlandse Mainports

Onder Druk, Spectrum: Speuren naar ontwikkelkracht Uitgeverij, Houten.Glasbergen, P. (2002) “The Green Polder Model: Institutionalizing Multi-

stakeholder Processes in Strategic Environmental Decision-making”, EuropeanEnvironment, 12 (6): 303–15.

Huxham, C. (2000) “The Challenge of Collaborative Government”, Public Manage-ment Review, 2: 337–57.

Jordan, G. (1990) “Sub-governments, Policy Communities and Networks: Refillingthe Old Bottles?”, Journal of Theoretical Politics, 2 (3): 319–38.

Keast, R., Brown, K. and Mandell, M. (2002) “Getting the Right Mix: UnpackingIntegration Meanings and Strategies”, International Public Management Journal,10 (1): 9–33.

Keast, R., Mandell, M., Brown, K. and Woolcock, G. (2005) “Network Structures:Working Differently and Changing Expectations”, Public Administration Review,64 (3): 363–71.

Ket, M. R. (2000) “Geluidshinderonderzoek en het Schipholbeleid”, in Bureau PAU,Schiphol en de Ruimtelijke Ordening: De ontwikkeling van het Schipholcomplex1980–2000, Groningen/Den Haag.

Kickert, W. J. M., Klijn, E. H. and Koppenjan, J. F. M. (eds) (1997) Managing ComplexNetworks: Strategies for the Public Sector, London: Sage.

Klijn, E. H. (2008) “It’s the Management Stupid!” On the Importance of Managementin Complex Policy Issues, The Hague: Lemma.

Koppenjan, J. F. M. and Klijn, E. H. (2004) Managing Uncertainties in Net-works: A Network Approach to Problem-solving and Decision-making, London:Routledge.

Kruijf, A. de (2002) Het Bulderbos: Verzet tegen uitbreiding van Schiphol 1993–2002,Amsterdam: Mets & Schilt.

Ministry of Physical Planning (1988) Fourth Memorandum on Spatial Development,Den Haag, 1988.

Ministry of Physical Planning (1995) PKB Schiphol en Omgeving, Deel 4, Den Haag.

Ministry of Transportation (1997) Integrale Beleidsvisie van het kabinet “Hoeveel ruimtegeeft Nederland aan luchtvaart?”, Den Haag.

Ministry of Transportation (1998) Strategische beleidskeuze toekomst luchtvaart,Den Haag.

Ministry of Transportation (1999) Toekomst Nederlandse Luchtvaart, Den Haag.Ministry of Transportation (2002) PKB Voortgang 2002. The Hague.Ministry of Transportation (2006) Eindrapport evaluatie Schipholbeleid, Den Haag.Ministry of Transportation, Physical Planning and Economic Affairs (2006) Cabinets

Perspective Schiphol, Den Haag.Nooteboom, B. (2002) Trust: Forms, Foundations, Functions, Failures and Figures,

Cheltenham: Edward Elgar.Olson, M. (1965) The Logic of Collective Action: Public Goods and the Theory of Groups,

Cambridge, Mass.: Harvard University Press.Ostrom, E. (1990) Governing the Commons: The Evolution of Institutions for Collective

Action, Cambridge: Cambridge University Press.O’Toole, L. J. (1988) “Strategies for Intergovernmental Management: Implement-

ing Programs in Interorganisational Networks”, Journal of Public Administration,25 (1): 43–57.

Parker. R. (2007) “Networked Governance or Just Networks? Local Governance of the Knowledge Economy in Limerick (Ireland) and Karlskrona (Sweden)”, Political Studies, 55: 113–32.

Parsons, W. (1995) Public Policy: An Introduction to the Theory and Practice of PolicyAnalysis, Cheltenham: Edward Elgar.

Pestman, P. (2001) “In het spoor van de Betuweroute”, PhD thesis, Amsterdam:Rozenburg Publishers.

Pierre, J. (ed.) (2000) Debating Governance: Authority, Steering and Democracy,Oxford: Oxford University Press.

Projectbureau Mainport and Milieu Schiphol (PMMS) (1991) Integrale versie Planvan Aanpak Schiphol en Omgeving. Den Haag: Sdu Uitgeverij.

Raad voor Ruimtelijk, Milieu–en Natuuronderzoek (RMNO) (2009) DuurzameOntwikkeling en Schiphol: Naar een creatieve confrontatie, Den Haag: RMNO.

Ragin, Charles C. (1992) “ ‘Casing’ and the process of social inquiry”, in Charles C.Ragin and Howard S. Becker (eds) What Is a Case? Exploring the Foundations ofSocial Inquiry, Cambridge: Cambridge University Press.

Rein, M. and Schön, D. (1992) “Reframing Policy Discourse”, in F. Fischer and J. Forester (eds) The Argumentative Turn in Policy Analysis and Planning, Durham,NC: Duke University Press.

Rhodes, R. A. W. (1990) “Policy Networks: A British Perspective”, Journal ofTheoretical Politics, 2 (3): 293–317.

Rhodes, R. A. W. (1997) Understanding Government, Buckingham: Open UniversityPress.

Richardson, J. J. (ed.) (1982) Policy Styles in Western Europe, London: Allen & Unwin.

Sabatier, P. (1988) “An Advocacy Coalition Model of Policy Change and the Roleof Policy-oriented Learning Therein”, Policy Sciences, 21: 129–68.

Scharpf, F. W. (1997) Games Real Actors Play: Actor-centered Institutionalism in PolicyResearch, Boulder, Colo.: Westview Press.

Soerenson. E. and Torfing, J. (eds) (2007) Theories of Democratic Network Governance,Basingstoke: Palgrave Macmillan.

Susskind, L. and Cruikshank, J. (1987) Breaking the Impasse: Consensual ApproachesResolving Public Disputes, New York: Basic Books.

Tan, G. (2001) Urgentie, leiderschap en kennisontwikkeling: verklaringen voor driebesluitvormingsronden over Schiphol tussen 1989 en 1991, Amsterdam: UitgeverijLemma.

Teisman, G. R. (2000) “Models for Research into Decision-making Processes: OnPhases, Streams and Decision-making Rounds”, Public Administration, 78: 937–56.

Termeer, C. J. A. M. and Koppenjan, J. F. M. (1997) “Managing Perceptions in Networks”, in W. J. M. Kickert, E. H. Klijn and J. F. M. Koppenjan (eds)Managing Complex Networks: Strategies for the Public Sector, London: Sage.

Tweede Kamer (1998) File 25466, Nr. 19. Letter from the Minister of Trans-portation to the Lower House.

Van Wijk, M. (2007) Airports as Cityports in the Cityregion. NetherlandsGeographical University Press.

Waarden, F. van (1999), “European Harmonization of National Regulatory Styles?”,in J. A. E. Vervaele et al. (eds) Compliance and Enforcement of European Com-munity Law, Deventer: Kluwer.

Weggeman, J. (2003) Controversiële besluitvorming, Utrecht: Lemma.

Werther, P. (1992) De uitbreiding van Schiphol: invloed op de besluitvorming van 1989tot 1991, Vrije Universiteit, Amsterdam.

Williamson, O. E. (1979) “Transaction Costs Economics: The Governance ofContractual Relations”, Journal of Law and Economics, 22 (2): 233–61.

Williamson, O. E. (1998) “Transaction Costs Economics: How it Works, where is itHeaded”, De Economist, 146 (1): 23–58.

GOVERNANCE, NETWORKS AND POLICY CHANGE:

THE CASE OF CANNABIS IN THE UNITED KINGDOM1

Beatriz Acevedo and Richard Common

Introduction

The two coterminous concepts of governance and policy networks have beenreasserted in public management studies in the wake of the recent decline inthe popularity of the so-called New Public Management (NPM) movement.The concepts appear to be inter-related with the notion of the policy net-work offering a perspective on efforts to “join up government”, which is alsoderived from “governance” approaches (DeLeon 2005). While this chapterfocuses on these concepts, it is clear that an understanding of networks isvital to the aspirations of the proponents of joined-up government in termsof achieving joined-up governance. It is also assumed that governance goesbeyond top-level policy coordination to where the public sector coordinatesand cooperates with nonstate actors, such as firms and voluntary organiza-tions, to deliver policy outcomes.

Policy networks have enjoyed acceptance and popularity amongst scholarsof public management for describing the complexity of the policy process inmodern states. Policy networks refer to the diversity of actors and relationshipsin the policy process beyond political bureaucratic relationships (Atkinsonand Coleman 1992: 156). Furthermore, the network concept has been definedand used in different ways. However, there is a broad agreement that it is ameso-level concept, which provides a link between the micro-level of analysis,which deals with the role of interests and government in relation to particu-lar policy decisions, and the macro-level of analysis, which is concerned withbroader questions regarding the distribution of power (Marsh and Rhodes1992: 1). The link with governance is that an understanding of networks provides a basis for horizontal approaches to service delivery and providesan alternative approach to that of NPM (Klijn 2005: 260).

This chapter examines the case of drugs policy in the United Kingdom in order to examine the usefulness of network approaches. Several studiesabout drugs regulations have focused on its social and historical develop-ments (Becker 1953, 1963; Berridge 1990, 2005; Mills 2000). Contemporaryresearch on drugs policy has evolved from the 1960s, adopting a more international perspective (Dorn et al. 1996; Boekhout van Solinge 2002) and analyzing drugs policy in the context of criminal policy in the UK(McLaughlin and Muncie 1994; Murji 1998). However, very few attemptsat analysis from a public management perspective have been made (Marlow1999). The fact that drugs policy is managed by a variety of actors at inter-national, state and local level, and that major tensions arise when agenciesare required to cooperate with each other while competing for resources, means that drugs policy has enormous potential for assessing the utility of network analysis. We examine these issues based on current research about drug policy in the United Kingdom, particularly the analysis of thedecisions around the reclassification of cannabis in the regulatory system ofdrugs during the last seven years. We use this example to provide practicaland empirical insights for the discussion about the utility of the enhancedpolicy network analysis proposed by Atkinson and Coleman (1992), in addition to developments in conceptualizing different types of networks as posited by Rhodes (1997). The chapter will also use this case to examinethe view of governance presented by networks. What is clear is that the rational or evaluative concerns of NPM cannot be applied to policy areassuch as drugs.

The chapter is divided as follows: in the first section, we discuss some of the theoretical and conceptual aspects of the main topics of this chapter:“policy networks”, “policy communities” and “governance”. In the secondsection, we present an overview of drugs policy in the United Kingdom, point-ing out the difficulties in regulating cannabis use. The application of policynetwork analysis to the case of cannabis reclassification is the subject of thethird section. In the fourth section we focus on the different “appreciations”of the cannabis problem emerging from the previous analysis of the policynetworks involved in the drugs policy discussion. We argue that the “appre-ciative system”, as proposed by Rhodes, combined with the differentialinfluence of certain policy networks, does determine decision-making indrugs policy. Taking into account these important considerations, we shallpresent some of the implications that such issues have for public-sector management, in particular the challenges that are faced in managing net-works of actors to achieve policy outcomes. Differing perceptions, and theimbalance of power within networks, present obstacles for the neat horizontalgovernance that prescriptive approaches to network management seem toimply (Richards and Smith 2002). Finally we offer some conclusions regardingthe utility of the central concepts of this chapter.

THE CASE OF CANNABIS IN THE UNITED KINGDOM

1. Policy networks and governance in the United Kingdom

Policy networks

Although beset by definitional squabbling, policy network analysis appearsto provide a useful empirical metaphor for understanding the complexity ofcontemporary policy processes:

The metaphor of a network or community seeks to focus on the pattern of formal and informal contacts and relationships, which shape policy agendas and decision-making as opposed to the inter-play within and formal policy making organizations and institutions.. . . Network analysis is based on the idea that a policy is framedwithin a context of relationships and dependences.

(Parsons 1995: 185)

In Britain the network metaphor has provided a framework to understandpolicy-making in democratic regimes. Richardson and Jordan (1979: 23) statedthat the policy-making map of Britain was characterized by a fragmentedcollection of subsystems – a “series of vertical compartments of segments,each segment inhabited by a different set of organized groups and generallyimpenetrable by ‘unrecognised groups’ or by the general public” (p. 74). Withthe increasing complexity of policy issues, the participation of networks indecision-making processes has defined a broad area to understand.

Rhodes has provided an extensive approach to network analysis, stress-ing the continuity in the relations between interest groups and governmentaldepartments (Rhodes 1988, 1997). Drawing on the European literature, hetakes a different approach from that of Richardson and Jordan, defining policy networks with regard to interorganizational relations rather than tosubgovernments, emphasizing the structural relationship between political institutions rather than interpersonal relations (Rhodes 1997: 36). Thus, hedefines a policy network as a cluster or complex of organizations connectedto one another by resources dependencies. The types of networks are to bedistinguished by their degree of integration, stability and exclusiveness. Inaddition, Rhodes identifies four dimensions along which networks vary: interests, membership, interdependence and resources (p. 39). In his typology,at one end of a continuum are policy communities that have stable andrestricted memberships, which are highly integrated within the policy-makingprocess. Issue networks, in contrast, represent a much looser set of interests,are less stable and are non-exclusive, and have much weaker points of entryinto actual policy-making. In the middle we have professional networks, inter-governmental networks and producer networks (pp. 37–8).

To enhance the advantages of using policy networks, Atkinson andColeman’s analysis suggests including three additional aspects: the macro

political context, the international dimension and the dynamic of politicalchange. In the first place, they question whether by disaggregating the statein different policy networks there would be little incentive to reconstitute the state or conceive of it as anything other than an assembly of organiza-tional actors. They also argue that the concept of a “policy network” hasbeen developed primarily to assess national policy-making from a domesticpoint of view. Whether the studies are comparative or limited to one country,the stress is on the identification of constellations of national or subnationalstate agencies, politicians, interest groups and political parties; they acknow-ledge that an increasing number of political areas are now determined at theinternational level (Rhodes 1997: 168).

Taking into account the complex nature of drugs policy, it is possible toevaluate the potential of using policy network analysis in understanding certain processes of policy-making involving multiple agencies. As drug policy making in the United Kingdom involves multiple actors, it is difficultto point to a dominant agency for either the formulation or the implementa-tion of related policies. Duke (2003) uses policy networks to analyze the relationship between prisons and drugs policy. She accounts for how differ-ent groups dispute authority over managing the drugs problem inside prisons,whilst emphasizing different values and appreciations concerning diverse institutions and groups involved in tackling this problem.

In addition to the multi-agency approach, a second element in this analysisis provided by the use of the notion of governance. Although it appears thatthe main characteristic of the drugs policy network corresponds to that ofan “issue network” in the classification of Marsh and Rhodes (1992), theattempt to steer networks in the managerialist interpretations of governancestill suggests the manipulation of targets and strategic planning to reach thepolicy outcomes formulated by central government. As Klijn (2005) pointsout, there are two managerialist interpretations of governance on offer. One is the NPM variant offered by Marsh and Rhodes (1992); the other isthe joined-up-government approach. Although Pollitt (2003a: 65–7) adds aconsiderable list of weaknesses to the network approach, he concedes thatwithin the context of governance it is capable of offering an analysis of inter-organizational dynamics.

Governance

Many contemporary debates about public management revolve around theconcept of governance. Much of its current attractiveness to both politiciansand policy-makers is that governance implies obtaining a much wider rangeof policy outcomes than those provided by the state. Although the term isnot new, governance has evolved from the shortcomings of NPM, involv-ing non-state actors in service delivery and the shift away from hierarchicalorganization in the public sector toward networks (Eliassen and Sitter

2008). In the case of many policy areas, including drugs, governance impliesa wider analysis of the outcomes delivered by policy that go beyond the mech-anistic assessments and the tinkering with inputs implied by the public management approach. In addition, the association with networks is clearowing to the centrality of multiple and often conflicting stakeholders at thecore of governance approaches to policy management.

Wälti et al. (2004: 86) emphasize the importance of “governance” in theiranalysis of Swiss drug policy:

“Governance”, “networks” and “partnerships” are widely used con-cepts to denote changes in the way public policies are managed. They capture transformation in the nature of coordination amongagencies and in their relationship with society. They imply that public action takes place in self-organizing networks that govern with-out recourse to hierarchy, or at least, with a significant degree ofautonomy from the state, and that public and private organizationscooperate and compete in a socio-cybernetic, horizontal and inter-organizational system of actors.

They highlight some aspects in the operation of policy networks in terms of actors and relationships that are useful for understanding the British case.For example, they present how media and social opinion play an importantrole in any decision on the drug policy, and how drug policy networks might be enhancing the representation of marginalized actors such as drugusers. A similar situation can be evidenced in the British discussion aboutcannabis policy, as will be analyzed in this chapter.

Clearly, there are limitations to political science interpretations of policynetworks under the rubric of “governance”. Rhodes (1997: 50) also recognizesthis when he identifies “governance as a socio-cybernetic system” as one possible use of “governance”. Although this term presents different inter-pretations, it can be assumed that it is in the interest of governance to keepa balance between maintaining control over political decisions and empower-ing policy networks by consultation, restitution or direct decentralization.

For this analysis, our first question is whether the object of governance isthe management of complex networks, which in theory reinforces subsystemautonomy enabling networks to resist central control; and autonomy isenhanced further because networks possess specialist knowledge and controlof “street-level” implementation (Taylor 2000: 52). Thus, the point here isto find out to what extent networks can effectively have autonomy to influ-ence policy-making and not just at the level of implementation or consultation.

Drugs policy represents an opportunity to analyze the extent of influenceof networks over policy formulation. In the case of cannabis, a number of networks have had representation in the “advisory councils” or the“commissions” in the different governmental agencies. They can be defined

as taskforces, and hence they may seem to be the “response of the Labourgovernment to the demands of central coordination of intergovernmental relations: the epitome of government steering rather than rowing” (Taylor2000: 53–4). With regard to this point, Atkinson and Coleman argue thatif policy networks exchange relationships, then the question would be: What is the structural context in which these exchanges take place? Are thererelationships of power and dependency that transcend and color individualtransactions?

If the answer is yes, then we will have to reach beyond transactionalanalysis to posit the structural conditions in which interaction occurs.. . . To bridge the gap it will be necessary to focus on institutionalvariables such as the level of centralization and professionalism thatcharacterize organizations in a network, and on ideological variables,such as the intellectual foundations of dominant worldviews in par-ticular policy areas.

(Atkinson and Coleman 1992: 157)

The next question is whether policy network analysis can provide a comprehensive framework for drug policy understanding by reviewing theevolution of the network approach. The analysis will also attempt to mapthe various vertical and horizontal interdependencies to emphasize theproblems of managing complex networks and to assess the extent to whichinterorganizational relationships are a necessary and potentially enduring feature of governance. To carry out this analysis, first we provide an over-view of drugs policy in the United Kingdom, focused on the case of cannabis.

2. Drugs policy in the United Kingdom

The development of drugs policy in Britain has been determined by a com-bination of, on the one hand, the different perceptions of the problem ofdrugs and, on the other hand, the international conventions regulating thesesubstances (Acevedo 2007). In the United Kingdom, the responsibility fordealing with the “drugs problems” has been disputed between the medicalprofession, who argue for a public health approach, and the criminal justicesystem, responsible for the application of the legislation and the adminis-tration of penalties for drugs offences (Berridge 1990; Stimson and Lart 2005;Bean 2002).

In particular, current UK drugs policy emerges as the product of bothdomestic discussions about the “problem of drugs” during the 1960s and the formalization of the international approach to narcotic drugs, articulatedin the Single International Convention of 1961. The result was formalizedby the Misuse of Drugs Act of 1971, which proposed a classification of drugsbased on their level of individual and social harmfulness; hence defining

different penalties for each substance. It divides drugs into three classes (Table 22.1).

The Misuse of Drugs Act of 1971 remains the most important output inUK drugs policy-making. The Act also placed the Advisory Council on theMisuse of Drugs (ACMD) as the relevant scientific body to consider any further legislative changes. Throughout nearly four decades there have been remarkably few changes, given that the drug market and the patternof consumption have changed dramatically. In contrast, new agencies haveappeared to address the drugs problem. However, it was only as recently as1995 that a comprehensive strategy was finally proposed. The initiative“Tackling Drugs Together” emanated from the Department of Health andseemed to apply a “multi-agency” approach. When Labour was elected in1997, they drew upon this strategy, emphasizing a “partnership” approachand acknowledging the experience gained in previous years. Therefore, it wasnot surprising that:

drugs were moved to the centre of the political stage. Drugs policywas to be “joined up” in the sense that a holistic multiagencyapproach was to be adopted. Intervention was to be “evidence-based”and closely monitored to ensure that it constitutes “best value”.

(Marlow 1999: 1)

Table 22.1 Classification of illegal drugs in the United Kingdom, based on theMisuse of Drugs Act, 1971

Penalties forsupply

Life imprisonmentor unlimited fine

Fourteen years’imprisonment orunlimited fine

Five years’imprisonment or fine

Penalties forpossession

Seven years’imprisonment orunlimited fine

Five years’imprisonment or unlimited fine

Two years’imprisonment or fine

Including(amongst others)

Cocaine, crack cocaine,ecstasy and relatedcompounds, heroin, LSD,“magic mushrooms” orpsilocybin, methadone,morphine and opium

Amphetamines, barbituratesand codeine Cannabis(herbal and resin) – after 29 January 2009 returnsfrom Class C to Class B.

Anabolic steroids,benzodiazepines andbupronorphine. Minortranquillisers are classifiedas C, yet possession is illegalwithout a prescription

A number of changes to this strategy were also implemented by the new Labour government, articulated in “Tackling Drugs Together: To Build aBetter Britain”. Amongst these changes, the government first insisted on across-cutting approach to drugs by appointing the UK Anti-Drug Coordin-ator, more popularly known as the Anti-Drug Tsar. Located in the CabinetOffice, the UKADC Unit was responsible for the coordination and imple-mentation of the drugs strategy (Blair 1998). The strategy was coordinatedand driven across government by the Cabinet Sub Committee on Drug Misuse,to which the Anti-Drugs Coordinator reported. Delivery at local level continued through Drug Action Teams (DATs) and their Drug ReferenceGroups (DRG), formed under the previous Conservative government. How-ever, this “joined-up” experience in coordinating drugs policy was rather short-lived: Labour’s second term of office saw the Home Office wrestingcontrol of drugs policy from the “cross-cutting” unit of the Cabinet Office,which was terminated.

In June 2001, the Home Office retook the task of managing drugs policywhile maintaining DATs. At the same time, the Home Office proposed a revision of cannabis policy by asking the ACMD for its opinion on the harm-ful aspects of cannabis. This development represents a milestone in UK drugs policy, and it is our main focus here. Indeed, despite a previous call for a revision of the status of cannabis in the legislation, the political context in 2001 provided a window of opportunity for policy change. A possible explanation is the role of policy networks and the growing popularity of governance approaches to so-called “wicked” social problems (Pollitt 2003:68). The set of questions posed by governance and public management inregard to other contemporaneous Labour initiatives such as “best value” and “joined-up government” influenced a traditional value-driven issue. Theresult was a policy revision called “What Is Working in the Government DrugsStrategy” (House of Commons 2002). On the other hand, it is possible tosay that this initiative was part of the Home Secretary’s strategy to regaincontrol over the drug issue, after the brief but high-profile tenure of the UK anti-drugs tsar. The process of cannabis reclassification appeared to bean extensive consultative process, taking into account the variety of actorsinvolved. Figure 22.1 aims to “visualize” the complexity of actors, networks,institutions and agencies involved in the policy-making debate.

3. Cannabis reclassifications: from B to C and back again

We use the case of cannabis reclassification as a representative example for our analysis of policy networks. In doing so, first we explain the processof downgrading and upgrading cannabis, and the role of different agencies,networks, institutions and actors in the public debate and in the policy-making process. The process of reclassifying cannabis and returning it againto its former class can be understood within changes to the Misuse of Drugs

should have decreased to seven years; however, penalties for supplying classC drugs were actually increased to fourteen years. At the implementationlevel, a wide range for discretion was also given to the police.

Following reclassification, wide-ranging commentary about cannabisbegan to appear in the media. One line of opinion was that the nature ofthe cannabis available has changed (dominated by a stronger variety called“skunk”), which was more likely to be detrimental to mental health. On the other hand, new scientific evidence contested the reclassification by suggesting that cannabis use was also linked to certain forms of psychosis (Castle and Murray 2004). Although this link is not new, the argument gainedprominence in the media and it added to the pressure against a relaxationon cannabis as perceived in relation to its downgrading.3 On the other hand,some official surveys suggest that the downgrading of cannabis in 2004 actually discouraged its use amongst young people rather than encouragedits consumption (Sian et al. 2007: 58). In 2008, in spite of this evidence and responding to increasing media pressure, the Home Office called for asecond round of consultations to evaluate this new evidence. The result ofthis review was contradictory: the ACMD (as the relevant body of expertsresponsible for evaluating the evidence) recommended keeping cannabis asclass C; the government took the decision of upgrading it back to class B.In the next section, we apply network analysis to explain this apparent policy reversal.

4. Applying policy network analysis

Although the concept of policy networks for analyzing drugs policy issuesis beset by definitional problems, drugs policy appears to display the charac-teristics of “issue networks”, which are unstable, have large numbers of members and “limited vertical interdependence” (Rhodes 1997: 38). Drugspolicy also appears as an “intergovernmental network” at the subnationallevel, particularly as it is based on an “extensive constellation of interests”.In the case of drugs, these constellations of interests are also articulated innational strategies, while at the same time there are international agenciesalso to be taken into account. Particularly in the case of the cannabisdebate, several types of policy networks participated. As defined by Rhodes,they are described in Table 22.2.

In Table 22.2, it seems that “issue networks” may well describe the pro-cess of policy-making in the case of cannabis. An issue network tends to havemany participants, fluctuating interactions, limited consensus, interaction basedon consultation, unequal power relationship and few resources (Rhodes 1997:45). In the case of cannabis reclassification, it can be demonstrated that thegreat variety of opinions mean that, although there is a remarkable processof consultation, this is not leading to a real negotiation, and politicians retaintheir discretion in final decisions which are unlikely to change core values

in the consideration of drugs. In fact, the analysis must include the distinctionbetween insider and outsider groups. Although the wide range of participantsand groups were incorporated in the discussion, at the end access to finaldecisions is only given to insider groups but only if they “fit” political think-ing on the issue (Dunleavy and O’Leary 1987: 163).

Another interesting aspect of this typology is the definition of a “producernetwork”; which would be crucial for an analysis of a legal economic activity.However, in the case of drugs, the presence of “producers” is unclear, sincethey remain illegal and organized by criminal gangs or “cottage industries”.Nonetheless, they are an important actor in the drugs economy, since home-grown cannabis provides approximately 50 percent of local consumption(Hough et al. 2003), and this trend has increased owing to the prolifera-tion of cannabis “factories” (Home Office 2008). In the United Kingdom,cannabis production does not present sufficient signs of cohesion or integrationto be defined as a “network”. In any case, the analogy is useful, as a kindof “consumer network” does exist, given the increased use of cannabis. How-ever, cannabis consumers do not have direct representation as a “consumernetwork” group, but their influence on drugs policy decisions should not be diminished by their “invisibility”.

Table 22.2 suggests that changes in drugs policy could have been influ-enced by different groups, to various degrees. Smith (1993: 51) argues thatpolicies that emanate from the state do not always reflect the demands ofgroups or classes but are the result of how state actors perceive their inter-ests, and how they perceive particular problems and their solutions. Duringthe discussion about cannabis reclassification from 2002 to 2009, several appre-ciations regarding the cannabis problem can be identified. Some networksdefine it as a matter of public policy in which the principles of NPM can be

Table 22.2 Policy networks involved in the policy-making on cannabis in theUnited Kingdom

Type of network

a. Policy community

b. Professional network

c. Intergovernmental networks

d. Producer networks

e. Issue networks

Participants

Police, Home Office, Health Department,criminal justice system

NHS, treatment system, medical doctors andpsychologists

Drug action teams

Not present in the debate, but active in thecannabis market

Activists, media, parents, campaigners, AdvisoryCouncil on the Misuse of Drugs (AMCD)

applied. This appreciation is characteristic of the original arguments support-ing cannabis reclassification, such as the House of Commons evaluation on drugs policy (House of Commons 2002), ACMD (2002, 2008) opinion, andan independent group of experts commissioned by the Police Foundation(Runciman 1999). On the other hand, cannabis is also defined as a matterof public health, emphasizing the risks associated with cannabis use involvingphysical and mental harm. In particular, “cannabis psychosis” has reinforcedmedical views on the problem of cannabis. Consequently, professional net-works of medical practitioners, psychologists and psychiatrists claim theirauthority on dealing with this issue. More generally, cannabis use has beenmainly defined as a matter of criminality. This view is supported by the generalperception of British society in which cannabis is associated with violenceand deviance, and reinforced by the number of news items pointing out thisconnection (Acevedo 2007). In terms of policy networks, this appreciationis expressed by policy community networks and intergovernmental networks,such as the Association of Chief Police Officers (ACPO 2003), the Home Office,the criminal justice system, and the International Narcotics Control Board.Finally, other networks, including activists and campaigners representing users,have claimed that cannabis use should be regarded as a “medicinal practice”,and for others it is a “recreational” practice (Parker et al. 1998). These viewshave been supported by reports commissioned by the government (i.e. ThePolice Foundation 1999; House of Lords Selected Committee on Science and Technology 1998, 2001). The public debate on cannabis reclassificationillustrates the coexistence of these different appreciations in defining the “problem of cannabis”. However, their existence as policy networks in ouranalysis does not guarantee that they are equally represented in the processof decision-making.

Atkinson and Coleman suggested that the identification of different policynetworks should be complemented by the consideration of wider issues. First,they recommend considering the issues of change and dynamism in politics.The case of cannabis evidences how the government’s interest and attentioncan change throughout time; thus, the initial impulse to the downgrading ofcannabis has been transformed in an actual opposition to the measure. Further,some interesting data have contested the government view about cannabisconsumption following the reclassification in 2004. For instance, figures for2007 show a decline in the level of consumption of this substance amongstthe British population. The proportion of 16–59-year-olds taking cannabishas decreased from 2002–3 when it reached 10.09 percent toward 8.2 per-cent in the latest British Crime Survey of 2006–7 (Sian et al. 2007: 58). Indeed,only 24.8 percent of young people aged between 16 and 24 years old tookcannabis during 2004–5 (after the reclassification of cannabis in January 2004) compared to 23.5 percent in 2003–4 within the same age group (Roe2005). Against this, it is possible to conclude that, although the consumptionis in decline, the visibility of the problem has provoked an increase of arrests

(Sian et al. 2007); hence the initial objective of diverting resources to classA drugs may not have been achieved.

In view of this evidence, decisions on cannabis policy should consider boththe facts and expert advice. However, as Atkinson and Coleman remind us,it is important to include the macro-political context. In this sense, consider-ing the volatile levels of popularity of New Labour, it seems that politicianswould not risk taking a stand against a rooted view on drugs as a matter ofcriminality and violence. In contrast to this view, the ACMD evaluated thegovernment idea of returning cannabis to class B by reviewing the evidence,consulting with different stakeholders, and calling a number of meetings andconferences. The result of this extensive process was for cannabis to remainas class C. However, the government managed to return cannabis to classB early in 2009. This reversal undermined expert opinion, since in this caseit seems the decision was taken based on the influence of a few “insider”groups rather than being informed by scientific advice or based on evidence.Thus, the very nature of an “evidence-based” policy, guided by expertadvice and scientific research, as proclaimed by NPM, is seriously questionedin this case.

Rhodes (1997: 39) offers an explanation by stating that “although decision-making within the organisation is constrained by other organisations, thedominant coalition retains some discretion. In this sense, it can be said thatpolitical communities tend to retain the power and they may act as those ‘dominant coalitions’ ”. With regard to this, Rhodes relates the idea inwhich the “appreciative system” of the dominant coalition does influence which relationships are seen as a problem and which resources will be sought. The notion of appreciation is crucial in understanding how certain problemsare defined, and thus how decisions are taken (Vickers 1995). Atkinson andColeman (1992: 174) observe that relevant knowledge may prevail when defin-ing political problems. Consequently, policy network analysis may benefitfrom a consideration of the diverse appreciative systems within different policy networks. It is also important to consider that expert opinion is notunified and different appreciations exist; this characteristic emphasizes theimportance of professional networks in defining drug problems. Moreover,Atkinson and Coleman consider the international context, which in the case of drugs policy includes institutions such as the International NarcoticsControl Board. The Board wields considerable power when approving or dis-approving any change in national policies concerning commitments acquiredby governments under international conventions (Acevedo 2007).

Although it is difficult to apply, we cannot dismiss the lingering NPM variant of governance when analyzing networks (Klijn 2005: 260). In thisinterpretation, different opinions should be taken by following a rational evaluation of the evidence and the advice of experts. However, this view ignoresthe power configuration and influences of some groups over others. In thisway, the role of the ACMD is crucial in the process of decision-making since

not only do they represent the expert advice but they are also responsiblefor the consultation process. Notwithstanding, the decision to return cannabisto class B contradicts this “rational” view and suggests that complexdynamics of power actually can challenge the logic of NPM. The case ofcannabis illustrates how the opinion of dominant groups inside the policynetworks can actually determine the development of the policy by disregardingevidence and expert advice in favor of decisions perceived as “popular”, suchas a hard policy on drugs. Furthermore, an analysis of drugs policy standsoutside market rationality, given the unorganized nature and criminality ofproducers and the diverse nature of consumers.

In synthesis, the case of cannabis policy evidences the importance of includ-ing the dimension of power in the process of policy-making. However, it isnot only a matter of influencing decisions but also of the way that problemsand subjects are constructed by the interaction of power and knowledge. Thusfar, we have shown how different appreciations of the problem of cannabisare shaped in the process of policy-making. Following Rhodes, these apprecia-tions determine how certain problems are defined in terms of public policy,and therefore influence the type of solutions, institutions and actions foraddressing that problem. Thus, it is possible to argue that changes proposedin the case of drugs policy will depend heavily on cultural values, institu-tional interests and prevailing discourses about the drugs problem. As a resultof this reflection, a challenging aspect for the policy networks analysis is theconsideration of power in relation to knowledge, and its relationship withfurther aspects of governmentality (Foucault 1991).

5. Implications for public management

The final aim of the chapter is to assess the utility of the network conceptfor public management. The case of drug policy, and particularly the recentexample of cannabis reclassification, shows the complexity of this issuewhen analyzing it through policy networks. Atkinson and Coleman’s (1992)critique of the network approach, as demonstrated in this chapter, providesimportant insights for enhancing the quality of policy network analysis. The rhetoric of NPM in Labour’s approach also seems to be tested in thecase of cannabis reclassification. Although originally the proposal was basedon the principles of value for money, efficiency in the use of resources, andevidence-based policies, this aim is hampered by the contrasting appreciationsof the drug problem. Hence, when policy networks discuss their appreciationsof the problem of cannabis, it must be clarified that they are not neces-sarily sharing the principles of NPM, but they are wrestling their own viewswithin the political context. This further demonstrates the fundamentalweaknesses of NPM in many policy areas.

As argued above, the results of the first phase of cannabis reclassifica-tion (2002–6) show a decrease in terms of cannabis consumption, indicating

a certain level of success of the strategy. Despite these results, the issue becamecontingent, and the government proposed the reversal on this measure.Ignoring the recommendation of the ACMD, and wishing to show a toughapproach, the government returned cannabis to class B. However, the pointhere is to analyze to what extent the advice of the experts is really taken intoaccount when dealing with “wicked” issues such as drugs policy. Followingpolicy network analysis it is clear that external networks, such as pressuregroups, mass media and so forth, although not necessarily responsible formaking political decisions, continue to play a crucial role in shaping “appre-ciations” of problems. Consequently, politicians tend to respond to these pressures rather than following expert advice or evidence-based information.

Clearly the implications for public management are very different fromthe NPM approach. The strategic role of the public manager horizontallyacross the network will defy the control structures wedded to the target-basedand performance management systems found across large swathes of the public sector. Network management (including partnerships) looks very different from the managerialist solutions of NPM. A normative literaturehas been developing around this theme for some time (for example, Lawton1999; Joyce 2000). The emphasis is on professional knowledge, managing relationships and understanding the nature of networks. The organizationalfragmentation and competition for resources under NPM would undermineany notion of a coherent drugs strategy.

However, the usefulness of the network approach is compromised by systems of policy evaluation that focus on outcomes, which are attributableto individual organizations. The ability of networks to learn by practice andthrough the acquisition of shared knowledge defies managerialist approachesto dealing with policy problems such as drugs. Institutionalized learning withinthe network may run contrary to the strategic intent of policy-makers as the network, in time, becomes highly resistant to change, forcing elected politicians either to “reinvent” policy or to “negotiate and persuade” throughnetworking (Pollitt 2003b: 47). A clear implication for public managementis to recognize and accommodate stakeholders, a core tenet of the govern-ance approach (Loffler 2009).

6. Conclusions

Policy network analysis has received both wide attention and criticism,although it remains of value as a heuristic device for analyzing complex policy processes. Drugs policy represents one of the more extreme cases, in which policy network analysis can be enhanced by introducing the threeaspects offered by Atkinson and Coleman: macro-political context, policychange and the international dimension. Following this framework, our analysis of drug policy shows that the dynamics of networks may be influ-enced by key state actors who have considerable command over resources

and thus dominate policy within the network. Macro-political aspects, in whichapparently pluralist systems may advocate for participation and extended consultation, remain dominated by insider groups and by an elite perspec-tive on public affairs. Nevertheless drugs remain a value-driven issue; thus,political decisions are likely to be influenced by public opinion, and the par-ticular action of stakeholders over a certain period of time. In this sense, thepresence of international pressure at the level of supranational structures,international agencies, and lobbyists and campaigners adds complexity tothis analysis.

In addition, governance principles advocate wider participation andempowerment of community partnerships. These principles, although diffi-cult to put into practice, are generating changes in the evaluation of drugspolicy, and they are actually influencing political decisions. The case of cannabisshows how governance approaches are supported by politicians, who tradi-tionally do not like to be seen as “soft on drugs”, in this particular case.Interactions between network members and participants, and where “stateboundaries” are broken down, tend to be fairly unique to particular networks,and drugs policy is no exception. An enduring feature of the governanceapproach is that it offers an explanation of network behavior to the publicmanager which contrasts sharply with the rationalist approach of NPM. More-over, Le Grand (2003) argues that governance rather than NPM-inspired systems of policy delivery are more likely to be robust in the long term. It is the same difference as between classical and processual approaches to strategic management.

Third, it is clear that managerialist approaches to policy analysis will be strictly limited within such a fluid environment where the drugs issue has such a high political saliency. However, it must be acknowledged that,whilst locating drugs as a problem to be managed, the debate may cut loosefrom a criminalized approach and its questionable results in time. Likewise,when emphasizing managerialism, some aspects related to the practicalitiesof drug policy may be incorporated into drug policy-making, in terms of alearning process. However, in some other cases it seems that managerialismmight continue to be an obstacle rather than a facilitator for practitionersand public officials in dealing with this complex issue.

In conclusion, we recommend complementing the analysis of policy net-works by stressing the importance of the “appreciative systems” within policynetworks, which are indeed determined by certain dynamics of power andknowledge (Foucault 1991). As demonstrated in this chapter, the rhetoric ofNPM in defining problems as a matter of rational management is contestedby the appreciations that different policy networks have about that par-ticular issue and their power in influencing final decisions. In this case, therole of media and social pressure is crucial in influencing the debate, whilepoliticians are driven by electoral purposes rather than by evidence-basedfigures or expert advice. In this context, the rationale of NPM within the

Labour approach is questioned by the “appreciative system” and the differentialpower of some actors and networks within the policy-making process.

1 This chapter is expanded and developed from B. Acevedo and R. Common (2006)“Governance and the Management of Networks in the Public Sector: DrugsPolicy in the United Kingdom and the Case of Cannabis Reclassification”, PublicManagement Review, 8 (3): 395–414.

2 A Guardian poll in 2002 showed that 53 percent of Britons opposed the measure.See A. Travis (2002) “Cannabis Relaxation Opposed by Majority”, The Guardian,31 July 2002. Available at http://www.guardian.co.uk/uk/2002/jul/31/drugsandalcohol.immigrationpolicy [accessed 15 January 2009].

3 For an extensive historical analysis of the origin of the link between cannabis andmadness in the context of colonial practices in India, please refer to J. Mills (2000)Madness, Cannabis and Colonialism, New York: St Martin’s Press.

References

Acevedo, B. (2007) “Understanding Cannabis Policy in the United Kingdom: A Systemsand a Post-structuralist Approach”, PhD thesis, University of Hull.

Advisory Council on the Misuse of Drugs (2002) The Classification of Cannabis underthe Misuse of Drugs Act 1971, London: Home Office.

Advisory Council on the Misuse of Drugs (2008) Cannabis: Classification and PublicHealth, available at http://drugs.homeoffice.gov.uk/publication-search/acmd/acmd-cannabis-report-2008 [accessed 29 March 2009].

Association of Chief Police Officers of England, Wales and Northern Ireland (2003)Cannabis Enforcement Guidance, London: ACPO.

Atkinson, M. and Coleman, W. (1992) “Policy Networks, Policy Communities andthe Problem of Governance”, Governance, 5 (2): 154–80.

Bean, P. (2002) Drugs and Crime, Uffculme: Willan.Becker, H. S. (1953) “Becoming a Marihuana User”, The American Journal of

Sociology, 59: 395–403.Becker, H. S. (1963) Outsiders: Studies in Sociology and Deviance, New York: The

Free Press.Berridge, V. (ed.) (1990) Drug Research and Policy in Britain, Aldershot: Gower.Berridge, V. (2005) “The ‘British System’ and Its History: Myth and Reality”, in

J. Strang and M. Gossop (eds) Heroin Addiction and the British System, Vol. 1,Origins and Evolution, Abingdon: Routledge.

Blair, T. (1998) “A Personal Statement from the Prime Minister”, in Tackling DrugsTogether: To Build a Better Britain, London: Home Office.

Boekhout van Solinge, T. (2002) Drugs and Decision Making in the European Union,Amsterdam: Mets & Schilt.

Castle, D. and Murray, R. (eds) (2004) Marijuana and Madness: Psychiatry andNeurobiology, Cambridge: Cambridge University Press.

DeLeon, L. (2005) “Public Management, Democracy and Politics”, in E. Ferlie, L. Lynn, Jr and C. Pollitt (eds) The Oxford Handbook of Public Management, Oxford: Oxford University Press.

Dorn, N., Jepsen, J. and Savona, E. E. (1996) European Drug Policies and Enforcement,London: Macmillan Press.

Duke, K. (2003) Drugs, Prisons and Policy Making, Basingstoke: Palgrave Macmillan.Dunleavy, P. and O’Leary, B. (1987) Theories of the State, Basingstoke: Macmillan.Eliassen, K. and Sitter, N. (2008) Understanding Public Management, London: Sage.Foucault, M. (1991) “On Governmentality”, in G. Burchell, C. Gordon and P. Miller

(eds) The Foucault Effect: Studies in Governmentality, Hemel Hempstead: HarvesterWheatsheaf.

Home Office (2008) Drugs: Protecting Families and Communities, available at http://drugs.homeoffice.gov.uk /publication-search /drug-strategy/drug-strategy-2008[accessed 29 March 2009].

Hough, M., Warburton, H., Few, B., May, T., Lan Ho, M., Witton, J. andTurnbull, P. (2003) A Growing Market: The Domestic Cultivation of Cannabis, York:Joseph Rowntree Foundation.

House of Commons, United Kingdom (2002) Third Report: The Government’s DrugPolicy: Is It working?, London: House of Commons.

House of Lords, United Kingdom (1998) Cannabis: The Scientific and MedicalEvidence, London: House of Lords Select Committee on Science and Technology.

House of Lords, United Kingdom (2001) Second Report: Therapeutic Uses of Cannabis,available at http://www.publications.parliament.uk/pa/ld200001/ldselect /ldsctech/50/5001.htm [accessed 15 January 2009].

International Narcotics Control Board (2002) Report of the International NarcoticsControl Board for 2002, Vienna: United Nations, available at http://www.incb.org/incb/annual_report_2002.html [accessed 15 January 2009].

Joyce, P. (2000) Strategy in the Public Sector, Chichester: John Wiley.Klijn, E. H. (2005) “Networks and Inter-Organizational Management: Challenging,

Steering, Evaluation, and the Role of Public Actors in Public Management”, in E. Ferlie, L. Lynn, Jr and C. Pollitt, C. (eds) The Oxford Handbook of PublicManagement, Oxford: Oxford University Press.

Lawton, A. (1999) “Managing Networks”, in A. Rose and A. Lawton (eds) PublicServices Management, Harlow: FT Prentice Hall.

Le Grand, J. (2003) Motivation, Agency and Public Policy, Oxford: Oxford Univer-sity Press.

Loffler, E. (2009) “Public Governance in a Network Society”, in T. Bovaird and E. Loffler (eds) Public Management and Governance, 2nd edn, Abingdon: Routledge.

McLaughlin, E. and Muncie, J. (1994) “Managing the Criminal Justice System”, in J. Clarke, A. Cochrane and E. McLaughlin (eds) Managing Social Policy,London: Sage.

Marlow, A. (1999) “Joined-up Thinking: Youth and Drugs Policy at the Millennium”,in A. Marlow and G. Pearson (eds) Young People, Drugs and Community Safety,Lyme Regis: Russell House.

Marsh, D. and Rhodes, R. A. W. (1992) Policy Networks in British Government, Oxford:Clarendon Press.

May, T., Warburton, H., Turnbull, P. J. and Hough, M. (2002) Times They Are a-Changing: Policing of Cannabis, York: Joseph Rowntree Foundation.

Mills, J. (2000) Madness, Cannabis and Colonialism, New York: St Martin’s Press.Mills, J. (2003) Cannabis Britannica: Empire Trade and Prohibition 1800–1928,

Oxford: Oxford University Press.

Murji, K. (1998) Policing Drugs, Aldershot: Ashgate.Parker, H., Aldridge, J. and Measham, H. (1998) Illegal Leisure: The Normalization

of Adolescent Recreational Drug Use, London: Routledge.Parsons, W. (1995) Public Policy: An Introduction to the Theory and Practice of Policy

Analysis, Aldershot: Edward Elgar.Pollitt, C. (2003a) The Essential Public Manager, Maidenhead: Open University Press.Pollitt, C. (2003b) “Joined-up Government: A Survey”, Political Studies Review,

1 (1): 34–9.Provan, K. and Milward, H. (1995) “A Preliminary Theory of Interorganizational

Network Effectiveness: A Comparative Study of Four Community Mental HealthSystems”, Administrative Science Quarterly, 40: 1–33.

Rhodes, R. A. W. (1988) Beyond Westminster and Whitehall: The Sub CentralGovernments of Britain, London: Unwin Hyman.

Richards, D. and Smith, M. (2002) Governance and Public Policy in the UK,Oxford: Oxford University Press.

Richardson, J. J. and Jordan, A. G. (1979) Governing under Pressure: The Policy Processin a Post-parliamentary Democracy, Oxford: Martin Robertson.

Roe, S. (2005) Drugs Misuse Declared: Findings from the 2004/05 British Crime SurveyEngland and Wales, London: Home Office Statistical Bulletin 16/05.

Runciman, R. (1999) Drugs and the Law: Report of the Independent Inquiry into theMisuse of Drugs Act 1971, London: The Police Foundation.

Sian, N., Kershaw, C. and Walker, A. (2007) Crime Survey England and Wales2006/2007, London: Home Office Statistical Bulletin.

Smith, M. (1993) Pressure, Power and Policy, Hemel Hampstead: HarvesterWheatsheaf.

Stimson, G. V. and Lart, R. (2005) “The Relationship between the State and LocalPractices in the Development of National Policy on Drugs between 1920–90”, inJ. Strang and M. Gossop (eds) Heroin Addiction and the British System, Vol. 1,Origins and Evolution, Abingdon: Routledge.

Taylor, A. (2000) “Hollowing out or Filling in? Task Forces and the Managementof Crosscutting Issues in British Government”, British Journal of Politics andInternational Relations, 2 (1): 46–71.

Vickers, G. (1995) The Art of Judgement: A Study of Policy Making, Thousand Oaks,Calif.: Sage.

Wälti, S., Kübler, D. and Papadopoulus, Y. (2004) “How Democratic Is ‘Governance’?Lessons from Swiss Drug Policy”, Governance: An International Journal of Policy,Administration and Institutions, 17 (1): 83–113.

CONCLUSIONS

Public governance and public services delivery: a research agenda for the future

Stephen P. Osborne

The introductory chapter to this volume explored the argument that publicpolicy implementation and public services delivery have evolved acrossthree regimes over the past century – from Public Administration (PA), throughthe New Public Management (NPM) and hence on to what has been termedhere “the New Public Governance” (NPG) in the contemporary state. Thesubsequent chapters of this volume have then explored the reality of this regimeshift in current public policy and public services delivery.

It is argued here that these chapters have demonstrated that public governance is indeed a significant paradigm for contemporary public servicesdelivery, embracing policy-making and a range of interorganizational andnetwork-based modes for public services delivery. It is also argued here thatthe NPG, by its very nature, requires a broader engagement with the environ-ment of public policy and public services than has perhaps been the case inthe past. This, in turn, requires greater attention to the issues of sustainability– not only in terms of public policies, public service organizations and publicservices themselves, but also in terms of the impact of these elements uponbroader issues of societal and environmental sustainability (Ball 2005).

It is debatable perhaps whether this is genuinely “new” – but public governance is certainly a reality in our modern fragmented public policy andpublic services delivery landscape. This concluding chapter now sets out anagenda to frame both public policy implementation and managerial practicewithin the NPG and “state of the art” research about its impact upon publicpolicy and public services delivery.

It is important at this stage to re-emphasize that the NPG regime is not being posed here either as a normative, policy alternative to PA and the NPM nor as “the one best way” (Alford and Hughes 2008) by which tomanage public policy implementation and public services delivery. Rather it is offered as a conceptual model that can both help us understand andevaluate the reality of public policy implementation and public services

delivery in the twenty-first century and clarify what the core challenges arefor public service managers within this “brave new world”.

It is also worth reiterating that the reality of “actually existing” public policy implementation and public services delivery is far more complex thana simple set of discrete implementation and delivery regimes or paradigms.In practice, these regimes will invariably coexist and interact rather than serially replace each other. Thus, it has been argued here that the NPG hasbecome the dominant regime of public policy implementation and public services delivery, with a premium being placed upon the development of sustainable public policies and public services and the governance of inter-organizational relationships. However, that does not take away either theimportance of the public policy process or the necessity to manage individualorganizational resources and performance in an efficient and effective man-ner. The world has simply become more complex, and we need to recognizethis in the conceptual models that we build. The bare bones of a more sophis-ticated approach are suggested, in a purely exploratory way, in Table 23.1.This explores the interplay of the policy and delivery regime and the focusof managerial activity.

This more sophisticated approach is perhaps a model to be developed further in the future. The intention here is more modest: to explore the newquestions that we need to start asking in order to understand and evaluatethe impact and import of the NPG for public policy implementation andpublic services delivery – and subsequently to use this knowledge to encour-age evidence-influenced public policy and managerial practice in the future.Accordingly, this chapter will consider the seven “new questions” that were articulated at the outset of this volume, as the core direction for futureresearch upon public policy implementation and public services delivery. These questions are:

CONCLUSIONS

Table 23.1 An exploratory model of the interaction of public policyimplementation and public services delivery regimes and managerial practice

Focus of managerial actionPolicy and service regime

Environment

Politicalmanagement

Competitivemarket behavior

Sustainable publicpolicy and services

Organization

Professional practice

Organizationalperformance

Boundary spanning andboundary maintenance

Policy

Street-levelbureaucracy

“Costs ofdemocracy”

Stakeholdermanagement

What should be our basic unit of analysis in exploring public policy imple-mentation and public services delivery – and what are the implications ofthis for theory and practice? (the fundamentals question)What organizational architecture is best-suited to delivering public servicesin the plural state? (the architectural question)How do we ensure sustainable public service systems – and what does sustainability mean? (the sustainability question)What values underpin public policy implementation and services delivery insuch systems? (the values question)What key skills are required for relational performance? (the relational skillsquestion)What is the nature of accountability in fragmented plural and pluralist systems? (the accountability question)How do you evaluate sustainability, accountability and relational performancewithin open natural public service delivery systems? (the evaluation question)

The new questions

The fundamentals question

Within PA, the unit of analysis was the public policy system as a closed system, with a focus upon the efficacy and impact of this system in its ownright.1 The key questions in this regime were about the effectiveness of thepolicy-making process and the extent to which public policy implementa-tion addressed the aspirations of the resultant extant public policies (see, forexample, Parsons 2003; Hill and Hupe 2009). A key concept here was thatof the implementation gap – concerned with why policy implementationmight fail. Often in this work, public managers were seen as “the villains ofthe piece”, who intentionally or unintentionally subverted these aspirations(Schofield 2001).

Within the NPM, the unit of analysis shifted to the individual public service organizations (PSOs) as open rational systems and to individual publicservice managers within these systems. It models public services delivery asan intraorganizational process that turns inputs into outputs (services) withina mediating environment, and with an emphasis upon the economy andefficiency of public services delivery. As noted in our introductory chapter,the NPM asked questions about the management and performance of suchorganizations and managers. Even when issues of interorganizational workingwere addressed, these were inevitably from the perspective of the individualorganization (for example, O’Toole et al. 2005).

Our argument is that such a discrete focus is no longer suitable for thestudy of contemporary public services. Rather a systemic approach is requiredthat views public services delivery from an open natural systems perspective.

CONCLUSIONS

This focuses attention upon the institutional and external environmental pressures that enable and constrain public policy implementation and thedelivery of public services within such a plural and pluralist system. Borrow-ing from the services management literature (for example, Gronroos 2007;Normann 2007), it is suggested here that the fundamental unit of analysisshould be the public service system. This includes not just the public policyprocess and PSOs (including their personnel and hard and soft service delivery technologies), but also the involvement of services users as the co-producers of public services (Pestoff et al. 2006) and the wider institutionaland environmental contingencies of public services delivery. Such an approachmoves beyond the concept of “simple” interorganizational networks as thefocus of attention for analysis, as in open rational systems. Rather it movesour attention to the inter-relationships between a number of interdependentelements of the public service system.

Co-production is a core element here. Public management theory has traditionally drawn upon theory from the business sector derived primarilyfrom manufacturing. However, there is a coherent body of theory about services management that may well be far more relevant to public servicesthan this manufacturing-derived theory. This body of theory is focused both upon this element of co-production in services and also upon the rela-tional nature of services that require governance rather than management(Gronroos 2007).

In this context, the key “fundamentals” questions for our research agendainclude:

• What can we learn from the services management literature that will helpto illuminate the nature of public delivery within the NPG regime orparadigm?

• What are the core elements of the service delivery system and how dothey inter-relate?

• What types of public service systems are there – and do they require a differential approach to their management and governance?

• What are the implications of the implicit element of user co-productionwithin public service systems?

• What is the role of PSOs and of networks of PSOs in delivering public services?

The architectural question

The architecture of public services delivery within PA was relatively straight-forward. This involved all elements of service provision being vertically integrated within PSOs – and with these bodies invariably being part of thegovernment sector. The structure and functioning of these governmental bodies was broadly similar for all services.

CONCLUSIONS

Within NPM, the architecture became more complex, with the plural provision of public services both from a range of PSOs and from a range of societal sectors (government, the third sector and the private sector). Thearchitecture of these organizations could vary in size, structure and functions,but was often broadly similar within societal sectors. Moreover, the marketacted as an isomorphic force to encourage homogeneity between organiza-tions providing similar services.

Such vertically integrated and sectoral approaches to organizationalarchitecture have broken down as the fragmentation and pluralism of theNPG has progressed. Increasingly the unit of analysis is not distinct organizations. Now it includes networks of both public managers and PSOsworking in concert to provide public services (O’Toole et al. 2007) and hybrid organizational forms that do not neatly fit within one distinctive societal sector and that have a fragmented and intermingled architecture (Evers 2008).

Now, the concept of hybrid organizations is not a new one. There has been a longstanding debate about “publicness” in PSOs and the blurring ofthe boundaries between public and private organizations. What is distinctivenow is the growing complexity of this hybridity. It is not simply a blurringof the boundaries between the public and the private. Rather it is the evolu-tion of genuinely new forms of organizational architecture that are not boundby sectoral limitations and that engender new forms of accountability.Social enterprises are one example of such hybrid organizations that mix socialends with a business orientation to income generation (Vidal 2008).

In this world, there is no “one best way” or prescribed approach to organ-izational architecture. It becomes a contingent process. Such new architectureneeds us to ask questions such as

• What organizational architecture is best suited to deliver what sorts of public services?

• What are the key contingencies of the contemporary architecture of PSOs,and what are their implications for public services delivery?

• How do PSOs develop organizational architecture that goes beyond simple organizational survival to tackle the issues of “boundary span-ning” and “boundary maintenance” in sophisticated service delivery systems that rely upon multiple elements for the successful delivery ofpublic services to local communities?

The sustainability question

To date, sustainability has often been viewed in public services delivery researcheither as concerned with the organizational ecology of PSOs and the sustain-ability of individual PSOs or as concerned with issues of environmentalism andecological sustainability. The contemporary architecture of public services,

CONCLUSIONS

as discussed above, requires a more sophisticated approach. This needs toconsider sustainability across a number of dimensions:

• The development of sustainable income streams and funding for publicservices, and the impact of these income streams upon their wider environment (perhaps through ethical investment strategies);

• The ecology of PSOs and their sustainability as discrete entities;• The sustainability of services delivered by PSOs and public service

delivery systems, and their impact upon their users and host com-munities; and

• The effect of public services delivery systems upon environmental andecological sustainability.

Some preliminary questions in this area are:

• What is sustainability? There is a need to move beyond the simpledefinitions of the Brundtland report.

• What are the key dimensions of sustainability for public service delivery systems? Are the ones suggested above the correct ones, or arethere alternatives or additional ones?

• To what extent is sustainability a marginal or mainstream issue for PSOs– and is it being “backwards incorporated” into existing service-led agen-das from the NPM paradigm?

• How has the concept and practice of sustainability evolved over the periodof PA–NPM–NPG?

The values question

The discussion about the role of values in public services has a long lineage.Within PA, there was an assumption of distinctiveness in the values of public-as opposed to private-sector management – and often an assumption of moralhegemony of the former. One of the first statements of the “public value thesis” was Sayre (1958); and the argument was developed by, among others,Rainey et al. (1976) and Murray (1975) – though the classic statement wasprobably Allison (1984), who argued that “public and private managementare at least as different as they are similar, and that the differences are atleast as important as the similarities” (p. 234).

As market and business disciplines came to have a greater influence onpublic management within the NPM, however, this hegemony of public values was challenged. A number of studies argued that public- and private-sector values and managerial practice were either not as differentiated as had earlier been argued or were converging (e.g. Posner and Schmidt 1996;Boyne 2002). Several studies also argued that public values had been replacedby new public-sector entrepreneurial values (Llewellyn et al. 2007), whilst

CONCLUSIONS

Mark Moore has coined the term “public value” as the equivalent of “share-holder value” within public organizations (Moore 1997).

Within the field of public governance, however, the debate about values hasbecome a contested one. Hoggett (2006: 192) has argued that PSOs now have

multiple tasks which are often in contradiction; they are certainlybeset by conflicting notions of what they should be doing and . . .for some organizations, paradoxically, it is important that they fail in order to maintain their contested legitimacy by serving thepublic’s unresolved ambivalence.

Because of this contested terrain in which PSOs now operate, and also becauseof the hybridity discussed above, public values have also become contested.Further, the evolution of the co-production of public services, as discussedin this volume by Pestoff and Brandsen, also poses new challenges both forthe way that service users are perceived by the staff of PSOs and for the valuebase of PSOs themselves. The competing discourses of “client”, “consumer”,“customer” and “citizen”, to name but a few, all imply different modes ofservice delivery for PSOs and a different set of values underpinning this service production.

For managers working in hybrid organizations, their values may need to span different societal sectors and be able to embrace the paradoxicalityof their new organizational architecture and logic. Thus, for example, Pooleet al. (2006) found both convergence and divergence in public- and private-sector managerial practice and values, whilst feminist critics have challengedthe public and private dichotomy as irrelevant in contemporary society(Nickel and Eikenberry 2006). As a consequence of these critiques, it is import-ant now to ask some fundamentally new questions about public values withinthe plural state and within public service systems:

• What values do public service managers and public service users hold,and how can the potential contradictions between these be governed?

• Does co-production require a distinctive set of values to underpin public services delivery – and what might the basis of these values be?

• What is the impact of contested values within public service systems uponthe delivery and use of these services?

• Do individual public services have distinctive values, and how can thesebe negotiated in complex service delivery systems?

The relational skills question

The development of the human resources2 of PSOs has undergone substantialtransformation over the last thirty years, and been subject to substantialresearch in its own right. The intention here is to describe the broad trajectory

CONCLUSIONS

of change over this period and to consider what questions we need to beasking now about the place and impact of human resources upon and withinthe NPG.

The approach to human resources under the dominant mode of PA waswell characterized by Farnham and Giles (1996: 118–19), reflecting upon thepractice prior to 1979:

Traditional people management practices and institutions of indus-trial relations in the public services had distinctive features and were broadly universalistic. These . . . were largely influenced bythe characteristics of the British state and government’s role as a “model” and “good practice” employer, assumed since at least 1917. This role originated from the state’s need to harmonise andimprove the effectiveness of public services provision nationally,enhance its political accountability and contain public expenditureas the public services expanded . . . the state’s objectives were to provide terms and conditions necessary to attract, retain and moti-vate the most skilled and professional staff, ensure harmonious andequitable employment practices across the public sector, and pro-mote stable employment relations.

This “model employer” role was first articulated by the Tomlin Commis-sion in 1929 and was subsequently refined by the Priestley Commission in1956. This emphasized the benefits of stability and continuity in employmentagainst those of higher rates of pay. Human resources were centrally con-trolled in this period, with pay and conditions often decided by central government and monitored through civil service departments. Local discre-tion was extremely limited (Farnham and Giles 1996).

As the NPM regime developed, one began to see changes in this approach.In its very nature, the NPM was antithetical to the “old” people manage-ment approach of PA. Increasingly staff became a human resource to be managed within the organization and with a strong emphasis upon the importa-tion of private-sector managerial and human resource practices into PSOs.Training began to be seen as a significant element of public services deliverywith an explicit focus upon the skills necessary for effective service deliveryby the organization (Farnham and Giles 1996).

Central to this approach to training and skills development was the competencies model that developed out of the work of David McClelland(1975). This approach sought to identify the distinctive skill competenciesrequired to undertake a specific task or post, and subsequently to use thesecompetencies as the basis for recruitment and ongoing training. This was subsequently developed by Boyatzis (1982) to specify twenty areas of corecompetencies for effective management across all sectors of the economy.Ironically, whilst the competency approach adopted within government

CONCLUSIONS

was intended both to try to match current resources to future needs and explicitly link these competencies to performance, the competencies adoptedwere invariably rooted within current, not future, practice. As Daley (2002)has noted in the US context, many competencies became fixed objectives andwere rarely re-evaluated in terms of the changing needs of PSOs.

This has become an especial problem as the issue of managing PSOs andpublic services in the fragmented state has evolved. Current competenciesinvariably emphasize the tasks of organizational and service management,and their links to performance. However, the reality of contemporary publicservice management is that it is an interorganizational and collaborative activ-ity, and requires the governance of complex systems and interorganizationalprocesses. Despite this, training has often remained rooted in organiza-tional needs rather than embracing the requirement to develop skills in managing the complex processes of interorganizational, network and systems governance.

The evidence base to begin this shift is growing. Within the private sector,the burgeoning literature on relational capital is making extant both the contribution of individuals to interorganizational working and the skills thatthey need to be effective in this (e.g. Kale et al. 2000). Some work has alsobegun to explore the competencies required for collaborative working (e.g.Huxham and Vangen 2005; Getha-Taylor 2008). Work is also needed on thenature of leadership within public service delivery systems, as opposed towithin individual PSOs (see Crosby and Bryson 2005; Mandell and Keast2009). However, this work is still at the margins of the field. In the main-stream, research continues to be dominated by concerns about the trainingand incentives required to enhance organizational performance. If our researchcommunity is to provide research that can meet the challenges of the newregime and provide a basis for evidence-influenced policy and practice, thenwe need to be asking new research questions about the skills required forrelational performance. These will include:

• What are the core elements of relational performance?• How do you identify, develop and sustain the relational capital of a PSO?• What key competencies are required for relational performance, and how

best can these be facilitated?• What are the links between such “boundary-spanning” activity and the

“boundary-maintenance” activity required for organizational survival – and what are the implications of this leadership within public servicedelivery systems?

The accountability question

Accountability is, and has been, a recurrent concern for public administrationand public management. It continues to be within public governance – but

CONCLUSIONS

the context has become that much more intricate. Traditional theories of public accountability with public administration were concerned with the formal lines of accountability, linking politicians and the political process tothe delivery of public services by public officials and governmental organ-izations (Day and Klein 1987).

Within public management, the discussion became far more complicated– now accountability concerned not just the accountability of politicians for public policy implementation but also the accountability relationshipsbetween governmental commissioners of public services and PSOs and theirmanagers based outside government – in both the commercial and the thirdsectors (e.g. Cutt and Murray 2000; Cribb 2006).

Public governance has brought these two issues together. By focusing onpublic service systems, it links both the role of politicians and political account-ability in these systems and the accountability of nongovernmental serviceproviders for the delivery of public services. It also links into the issue ofsustainability discussed earlier and the need for accounting systems that gobeyond traditional formal accounting approaches and focus (e.g. Gray 2002).Key questions here now include:

• What is the focus of accountability?• How do you establish accountability in fragmented public service delivery

systems – how can you hold interorganizational systems and networksaccountable for the delivery of public services?

• How can you build sustainability into the accounting process?• What is the interaction between the value base of public services, the

relationship of these services to their users (and/or co-producers) andthe accountability required by these services?

A key issue for the implementation of accountability in a fragmented stateis discussed further below.

The evaluation question

In many respects, performance evaluation and management are a true child of the NPM. It would be a gross overstatement to say that serviceswere never evaluated within the PA regime. They clearly were. However, the framework was one of the hegemony of the professional paradigm – services were evaluated against professional standards and service objec-tives, and rather more rarely against resources or strategy (Broadbent andLaughlin 2002).

As Hood (1991) has made clear, though, performance management andperformance metrics were at the heart of the NPM. Subsequently, a wholegeneration of researchers around the world have developed significant researchagendas exploring this issue (e.g. Boyne 1999; Boyne et al. 2003). The core

CONCLUSIONS

of this approach to the management of public services and of PSOs has been their conceptualization as open rational systems that process resourceswithin a constrained environment in order to produce public services. Import-antly, the focus has been upon PSOs as the unit of analysis rather than uponmore complex service delivery networks or systems.

The classic model of this approach has to be the “production of welfare”model developed by Knapp (1984). Highly innovative in its time, this modelexplored the production of social care services precisely as such an open rational system, where inputs were transformed into service outputs and produced outcomes for their users. Interestingly, Knapp argued againstattempts to try to evaluate outcomes for service users. This was in part becausethe timescales required to do so properly were too long to be useful for managing public services in real time. However, it was also because of thedifficulty of determining whether a particular outcome was directly attri-butable to a specific service, owing to the effect both of the intervening variables in the environment and of the intrinsic characteristics of serviceusers themselves (see, for example, DoH 1991).

Partly because of the influence of Knapp’s work and partly because serviceoutputs lent themselves more easily to the sort of quantitative analysis thatbegan to fuel public policy-making in the 1980s and beyond, the “accountinglogic” (Broadbent and Laughlin 2002) of performance management becamedeeply embedded within public services management as part of the NPMregime. It drove both the strategic planning process within PSOs and theiraccountability to local and central government for delivering public servicesto the local community.

It can be argued that this performance management culture is one of thelasting legacies of the NPM – and, indeed, in some countries (for example,England) this has been extended over subsequent years from the “straight-forward” use of indicators to evaluate organizational and service performanceand toward the use of performance targets and indicators as a way to con-trol the delivery of public services. They have become part of a regulatory,rather than simply a performance-oriented, state (Jayasuriya 2004). For theresearch community, this effect is seen most strongly in the recent ESRC PublicServices Programme, which has had at its heart the whole issue of organiza-tional and service performance management and performance metrics.

Our argument is that this fixation upon the metrics of organizational and service performance is now at odds with the reality of public servicesdelivery in contemporary society. Again, as has been emphasized several timesbefore on other elements of the service system, this is not to say that suchperformance evaluation of organizations and services is now irrelevant. Itmanifestly is not – and individual PSOs and public services will always needto monitor and evaluate their performance against their available resourcesand their strategic objectives. However, the reality of public service systems,as discussed above, has moved on from the “service silos” of the 1980s and

CONCLUSIONS

1990s. Performance management and performance metrics have not movedon with them, but rather have remained locked within the open rational sys-tems of the NPM. To take one example, efforts to use performance metricsto integrate sustainability into the core of public services management haveto date been thwarted by the resilience of the service focus of local govern-ment in England. Rather sustainability has been reinterpreted and backwardlyintegrated into existing performance metrics, emasculating its potential fortransformational change (Ball 2005).

This is not to say that there are not already models of performance management and accounting that could drive forward such a new approach.The social accounting tradition (Gray 2002) has the potential to reframe theapproach to performance within public services to include broader issues thanservice and financial performance alone. It has certainly begun to influenceperformance evaluation in some parts of the world (for example Italy – see Marcuccio and Steccolini 2005), whilst some have also argued stronglyfor its potential to advance “an agenda for the social justice dimension ofsustainability” (Ball and Seal 2005: 471). Elsewhere, work has also begun on evaluating the performance of collaborative networks, rather than of theconstituent organizations (Head 2008; Mandell and Keast 2008).

There are thus a raft of new issues that need to be addressed in using performance management to address the new challenges of the delivery ofpublic services within the NPG regime. These are issues located within anopen natural systems approach discussed previously, which emphasizes theinter-relationship of the organization and its environment, the fragmenta-tion of services delivery within contemporary society, the primacy of systemslogic and the pre-eminence of values. It is also necessary to move beyondthe simple (?) evaluation of individual PSOs and to ask some fundamentalquestions about how you evaluate multi-organizational performance and co-production in public service delivery systems. These questions include:

• How do you evaluate performance of a multi-organizational service system, as opposed to a single organization or service, and which systemincludes service users as co-producers?

• How do you evaluate “relational performance” in the sense of the performance of collectivities of PSOs and service users that co-producepublic services within service delivery systems?

• How do you evaluate the effectiveness of PSOs in working together in terms of both their ability to create and sustain relational capital withintheir organizations and to work effectively with other PSOs (that is, theevaluation of the process of interorganizational working, rather than itsoutputs or outcomes)?

• Is it possible to evaluate outcomes in a meaningful way that providesinformation that can be used in the planning and management of public services?

CONCLUSIONS

• How do you actually evaluate the sustainability of public services and PSOsand their contribution to societal sustainability in its widest sense – andin a way that allows transformational change in service delivery systems?

• How do you integrate public values into performance evaluation so thatissues such as equity and social justice can be part of the judgment ofservice systems, as well as economic performance?

• How can you use performance metrics to open up the debate about contestability within public services – in terms of service goals, valuesand impacts?

Conclusions

This volume began by questioning the extent to which a “new public govern-ance” exists or not. The evidence presented here, notwithstanding Hughes’shealthy skepticism, is that public governance does indeed exist and is a definingelement of public service delivery systems in the twenty-first century. Its meaning may continue to be contested; and, at its worst, it can mean “allthings to all people”. Nonetheless the concept does capture the challengesof the delivery of public services within fragmented service delivery systemsaround the world. The challenges may not be the same in all parts of theglobe, but relational values are at their core.

This chapter has suggested that a new research agenda is required to capture fully the complexity and diversity of public governance in this globalcontext. It is an ambitious one. However, the argument here is that it is onethat must be embraced if we are to move forward and consider the realitiesof public policy implementation and public services delivery within publicgovernance. Again, this is a not a “boosterist” agenda that argues for theNPG as the “one best way” to implement policy and deliver public services.Nor does it suggest that the need for effective public policy processes andfor effective organizational management has gone away. Rather it asks howthese challenges can be approached within the fragmented, interorganizationaland contested space that now comprises “the public sector”.

This concluding discussion has suggested what these new questions mightbe. It is certainly not intended to be either exhaustive or conclusive. Ratherit is intended as a contribution to fire a debate about the research agendaon public governance for the next decade and beyond. The research com-munity must embrace this new research agenda if it is to stop asking “oldquestions” that do not reflect the reality of public services in the twenty-firstcentury. The “new questions” suggested here do not, of course, negate theneed to continue exploring some of these “old questions” – but rather to do so in a context that reflects the new context of public governance. These“new questions” are ones that must be addressed in order to drive forwardevidence-influenced public policy implementation and public services deliveryin the twenty-first century.

CONCLUSIONS

Notes1 For a broad discussion of systems approaches see Scott (1992).2 The appropriate term to use here is also a subject of dispute, with each variation

having its own discourse. Thus, for example, you will find the field described variously as “human resource management”, “people management”, “personnelmanagement” and, perhaps more narrowly, “workforce planning”.

ReferencesAlford, J. and Hughes, O. (2008) “Public Value Pragmatism as the Next Phase of

Public Management”, American Review of Public Administration, 38 (2): 130–48.Allison, G. (1984) “Public and Private Administrative Leadership: Are They Funda-

mentally Alike in All Unimportant Respects?”, in Leadership and OrganizationalCulture: 214–39.

Ball, A. (2005) “Environmental Accounting and Change in UK Local Government”,Accounting, Auditing and Accountability Journal, 18 (3): 346–73.

Ball, A. and Seal, W. (2005) “Social Justice in a Cold Climate: Could SocialAccounting Make a Difference?”, Accounting Forum, 29: 455–73.

Boyatzis, B. (1982) The Competent Manager, New York: John Wiley.Boyne, G. (1999) “Processes, Performance and Best Value in Local Government”,

Local Government Studies, 29 (3): 1–15.Boyne, G. (2002) “Public and Private Management: What’s the Difference?”,

Journal of Management Studies, 39 (10): 98–122.Boyne, G., Farrell, C., Law, J., Powell, M. and Walker, R. (2003) Evaluating Public

Management Reforms, Buckingham: Open University Press.Broadbent, J. and Laughlin, R. (2002) “Public Service Professionals and the New

Public Management”, in K. McLaughlin, S. Osborne and E. Ferlie (eds) New PublicManagement: Current Trends and Future Prospects, London: Routledge.

Cribb, J. (2006) Being Accountable, Wellington: Institute of Policy Studies.Crosby, B. and Bryson, J. (2005) Leadership for the Common Good, San Francisco,

Calif.: Jossey-Bass.Cutt, J. and Murray, V. (2000) Accountability and Effectiveness Evaluation in Non-

profit Organizations, London: Routledge.Daley, D. (2002) Strategic Human Resource Management, Upper Saddle River, NJ:

Prentice Hall.Day, P. and Klein, R. (1987) Accountabilities, London: Tavistock.Department of Health (DoH) (1991) Assessing Outcomes in Child Care, London:

The Stationery Office.Evers, A. (2008) “Hybrid Organisations: Background, Concepts, Challenges”, in

S. Osborne (ed.) The Third Sector in Europe, London: Routledge.Farnham, D. and Giles, L. (1996) “People Management and Employment Relations”,

in T. Sergiovanni and J. Corbally (eds) Managing in the New Public Services, published by University of Illinois Press, Chicago p. 425.

Getha-Taylor, H. (2008) “Identifying Collaborative Competencies”, Review of PublicPersonnel Administration, 28 (20): 103–19.

Gray, R. (2002) “The Social Accounting Project and ‘Accounting Organizations andSociety’: Privileging Engagement, Imaginings, New Accountings and Pragmatismover Critique?” Accounting Organizations and Society, 27 (7): 687–707.

CONCLUSIONS

Gronroos, C. (2007) Service Management and Marketing, Chichester: John Wiley.Head, B. (2008) “Assessing Network-based Collaborations: Effectiveness for Whom?”,

Public Management Review, 10 (6): 733–50.Hill, M. and Hupe, P. (2009) Implementing Public Policy, London: Sage.Hoggett, P. (2006) “Conflict, Ambivalence and the Contested Purpose of Public

Organizations”, Human Relations, 59 (2): 175–94.Hood, C. (1991) “A Public Management for All Seasons?”, Public Administration,

69 (1): 3–19.Huxham, C. and Vangen, S. (2005) Managing to Collaborate, London: Routledge.Jayasuriya, K. (2004) “The New Regulatory State and Relational Capital”, Policy

and Politics, 32 (4): 487–501.Kale, P., Singh, H. and Perlmutter, H. (2000) “Learning and Protection of

Proprietary Assets in Strategic Alliances: Building Relational Capital”, StrategicManagement Journal, 21 (3): 217–37.

Knapp, M. (1984) The Economics of Social Care, Basingstoke: Macmillan.Llewellyn, N., Lewis, P. and Woods, A. (2007) “Public Management and the

Expansion of the Entrepreneurial Ethos?”, Public Management Review, 9 (2):253–68.

McClelland, D. (1975) A Competency Model for Human Resource Management Spe-cialists to Be Used in the Delivery of the Human Resource Management Cycle, Boston,Mass.: McBer.

Mandell, R. and Keast, R. (2008) “Evaluating the Effectiveness of Inter-organizationalRelations through Networks: Developing a Framework for Revised PerformanceMeasures”, Public Management Review, 10 (6): 715–32.

Mandell, M. and Keast, R. (2009) “A New Look at Leadership in CollaborativeNetworks”, in J. Raffel, P. Leisink and A. Middlebrooks (eds) Public SectorLeadership, Cheltenham: Edward Elgar.

Marcuccio, M. and Steccolini, I. (2005) “Social and Environmental Reporting in LocalAuthorities: A New Italian Fashion?”, Public Management Review, 7 (2): 155–76.

Moore, M. (1997) Creating Public Value: Strategic Management in Government,Cambridge, Mass.: Harvard University Press.

Murray, M. (1975) “Comparing Public and Private Management: An ExploratoryEssay”, Public Administration Review, July–August: 364–71.

Nickel, P. and Eikenberry, A. (2006) “Beyond Public vs. Private: The Trans-formative Potential of Democratic Feminist Management”, Administrative Theoryand Praxis, 28 (3): 359–80.

Normann, R. (2007) Services Management, Chichester: John Wiley.O’Toole, L., Meier, K. and Nicholson-Crotty, S. (2005) “Managing Upward, Down-

ward and Outward: Networks, Hierarchical Relationships and Performance”,Public Management Review, 7 (1): 45–68.

O’Toole, L., Walker, R., Meier, K. and Boyne, G. (2007) “Networking in Com-parative Context”, Public Management Review, 9 (30): 401–20.

Parsons, W. (1995) Public Policy, Cheltenham: Edward Elgar.Pestoff, V., Osborne, S. and Brandsen, T. (2006) “Patterns of Co-production in

Public Services”, Public Management Review, 8 (4): 591–6.Poole, M., Mansfield, R. and Gould-Williams, J. (2006) “Public and Private Sector

Managers over 20 Years: A Test of the ‘Convergence Thesis’ ”, Public Adminis-tration, 84 (4): 1051–76.

CONCLUSIONS

Posner, B. and Schmidt, W. (1996) “The Values of Business and Federal GovernmentExecutives: More Different than Alike”, Public Personnel Management, 25 (3):277–89.

Rainey, H., Backoff, R. and Levine, C. (1976) “Comparing Public and PrivateOrganizations”, Public Administration Review, March–April: 233–44.

Sayre, W. (1958) “Premises of Public Administration”, Public Administration Review,18 (2): 102–5.

Schofield, J. (2001) “Time for a Revival? Public Policy Implementation”, Inter-national Journal of Management Reviews, 3 (3): 245–63.

Scott, W. (1992) Organizations: Rational, Natural and Open Systems, Upper SaddleRiver, NJ: Wiley/Prentice Hall.

Vidal, I. (2008) “The Role of Social Enterprises in Europe: A Core Element or aDistraction in the Provision of Public Services?”, in S. Osborne (ed.) The ThirdSector in Europe, London: Routledge.

CONCLUSIONS

Coordination 42Co-production 227–8, 279–81, 416Corporate Governance Model 258Culture 169–72

Decision Making 41Diversity 73–4, 84Dynamics 73–4

Economism 23Efficiency 133, 134European Union (EU) 46–7, 114,

127, 142, 152, 156, 242, 358Evaluation 422–5

Flexibility 171Four Governance Model 257–60,

262–3

Globalization 114–15Governability 74–6, 78–81, 89Governance 6–7, 19–21, 31–2, 37–8,

41, 43, 72, 87, 102, 105, 111, 118,155, 394–5, 397–9

Governance bureaucratic 36Governance, co- 83Governance, collaborative 81–3Governance, corporate 91, 93–4, 102,

303–4Governance, definitions of 87–8, 142,

303–5Governance, global 20, 21–4, 24–7,

27, 31Governance, hierarchal 83Governance, interactive 73–4, 80,

61–4, 307–8Governance, local 341–5Governance, new 38

Accountability 42–3, 133, 135, 143,171–2, 282–87, 421–22

Accounting logic 423Accounting, social 424Administrative reform 173–4Advocacy coalitions 353, 357–9, 366Autonomy 171–2

Budgets 45–6

Capacity building 133, 136, 137, 295– 6

Capital, human 250Capital, relational 184, 193– 4, 259Civil society 22, 23, 49, 72, 79, 109,

116, 142, 274, 286Competition 287–91, 292Congestion charging 60–1Collaboration, aims 163–6Collaboration, cross-sector 202–5Collaborative advantage 163– 4,

178, 180–1Collaborative inertia 163– 4, 177–8Collaborative success 178–81Co-location 140Commodity 72–6Community organizations 138Community participation 133, 137–8Complexity 73–4, 153–4, 172Contract governance 7, 154Contracts 243Contracting 46, 57–9, 128, 150, 154,

155, 156–7, 240–2, 270–3, 273–7,277–95

Contracting, performance 277–82, 283Contracting, politics of 291–5Contractualism 143–4, 156Convergence 116

Governance (NPG), New Public 1–2,6–7, 9, 10, 30–32, 97, 106–7,109–13, 114–17, 118–19, 142–4, 150,152–3, 191, 322–3, 331–2, 413–15,424–5, 475–6

Governance, network (networked)259–60, 305, 306–9, 365– 6, 388–9

Governance, order of 80–1Governance, self 83Governing 55Governing system 76, 78–81Government 55, 56, 66–7, 68, 89–90,

95, 98

Hierarchy 38, 44Hybridization 79, 109, 417, 419

Infrastructure networks 150–2, 158–9

Innovation 52–3, 54–5, 62, 63–4, 131,137, 227–8, 231–3, 309, 315

Innovation, ancillary 54Innovation, public sector 55–6Innovation system 54Institutional independence 92Institutional relationships 72, 81Institutions 105–6, 113, 115,

141–2, 386Interactions 72, 74, 75International Labour Organisation

(ILO) 22International Monetary Fund (IMF)

22, 23, 29Inter-agency cooperation 134–8Interaction order 170–1Interdependency 141Iron triangle 353–4Isomorphism 116, 117, 232Issue networks 403–7

Joined up government 23, 343–4

Keynesianism 31Knowledge transfer (exchange) 174–7,

177–8, 314–5

Law 46–7, 119Leadership 139, 176–8, 200–1, 201–2,

210–11, 217–18, 219, 220, 344Local Area Agreements 345–6Local government 338–9, 339–41Local Strategic Partnerships 344–5, 346

Market Governance Model 258–9Market, the 49, 79, 97–101, 339–40Market failure 243Marketing 183–4Marketing, de- 184Marketing, public sector 184–5, 187–9Marketing, relationship 184, 190,

191–4Marketing, social 184Marketing, transactional 187, 189, 190Monopoly 244–5Monopsomy 245–7Meta-governance 37–8, 40, 43–8, 48–9Meta-governance, institutions of 44–8Multi-actor perspective 24–7

New City Management 113New competition 256New federalism 256New institutionalism 187New Public Governance (NPG) see

Governance, New PublicNew Public Management (NPM) 1–2,

3–4, 8, 10, 23, 39–40, 41, 42, 43, 44, 46, 91, 93, 94–7, 102, 105–6,109–111, 142–4, 159, 270–3, 286–7,304, 337, 339–41, 394, 407–8,413–14, 415, 417, 420–1, 422–4

New Public Service 113Network Governance Model 259–60Networks 25, 26, 27, 40, 41, 42, 48,

64–5, 91, 93, 97–101, 116, 140,316–7, 317, 324, 329–30

Networks, issue 354–6Non-Governmental Organizations

(NGOs) 22, 27, 40, 272

Organisation for Economic Co-operation and Development(OECD) 22, 89–90, 128, 149, 150, 152

Organisational boundary 64–5, 69Otherness 170Outcomes 142Outsourcing 248–52, 255–9, 261

Participation 41–2, 82Partnership 59–60, 65–6, 127–8, 143,

345–7Partnership, benefits of 130–4Partnership, definition of 128–30Partnership, problems of 134–8

Path Dependency 106Performance implementation 327–9Performance management 44–5,

249, 329Plural State 7Policy committees 356–7Policy coordination 45Policy games 366, 367, 384–5, 385–7Policy implementation 5, 202, 323–4,

324–7Policy Network Analysis 403–7,

408–10Policy networks 351–2, 366–8, 394,

396–7Policy networks, definition of 352–3Power relations 137, 168–9, 202, 204,

385–6Private Finance Initiative (PFI) 129,

143, 49Private sector 150–2, 153, 154–5, 287–9Privatization 150, 154, 158–9, 239Procedural Governance Model 258Production of Welfare Model 423Professionalisation 278– 80Public Administration 1–2, 2–3, 8, 10,

107–9, 112–13, 114–17, 118–19Public goods 63Public policy 20, 68Public governance see Governance,

New PublicPublic Private Partnerships (PPPs)

128, 129, 143, 149–50, 163, 194, 256, 322–3

Public Service Organizations (PSOs) 4,255

Public value 59, 68, 204

Reciprocity 116, 141Reform 36–7, 38–40, 48–9Regulation 157, 183, 282– 7, 288Relational skills 419–21Relationship management 193, 294–5

Representative democracy 41–2Resource Dependency Theory 187Resources, financial 65–6, 131–2,

134–5Risk 150, 155–6, 167–8, 192, 260–2,

309–12

Service integration 132–3, 247–8Service Level Agreements 136–7State, the 21, 25, 49, 79, 91, 99, 118,

138, 142Strategic management 45–8, 139, 144,

154–5, 189, 260–3Strategic positioning 189Supra-national Institutions 30–31Systems (also Systems Theory) 74–5,

76, 76–7, 415–6Systems Socio-cybernetic 91, 92, 102Sustainability 417–8

Third (also voluntary) sector 132,137–8, 223–4, 225–8, 270–3, 290

Third Way 23Transaction costs 155, 158, 187, 311,

313–4, 315, 391Transparency 251–2Trans-national policy 27Trans-nationalism 21, 26Trust 47–8, 139–40, 168–71, 190,

192–3, 204–5, 305–8, 309–12, 313–15, 370–71

United Nations Educational, Cultural and Scientific Organisation (UNESCO) 22

United Nations (UN) 28

Value, public 48, 418–9Value, claims 78Values 47–8

World Bank 92

Stephen P. Osborne Editor the New Public Governance Emerging Perspectives on the Theory and Practice...

Documents

Public Service Media Governance

GOVERNANCE & PUBLIC POLICY INITIATIVE

Simon Osborne - The Value of Good Corporate Governance

Public Governance in the Public Sector: Literature review

AVID in Public Governance

Public Governance in Lebanon

Public Governance and Territorial Development Directorate · Public Governance and Territorial Development ... Public Governance and Territorial Development Directorate ... or of

Good governance in public land management · Goals, progress in good governance and transparent ﬁscal management of the public sector. Good governance in managing public land ﬁrst

Chapter 10. Public Governance

PUBLIC GOVERNANCE REVIEWS - OECD Governance Reviews... · 2016. 3. 29. · PUBLIC GOVERNANCE REVIEWS OECD Public Governance Reviews were launched in 2007, to provide governments with

Governance - Public Service

Public Administration as Governance

DEMOCRATIC GOVERNANCE AND PUBLIC ......Democratic governance and public administration in the Western Balkans – general assessment Improvements to democratic governance remain crucial

A New Theory for Public © The Author(s) 2012 Service ... · 136 American Review of Public Administration 43(2) New Public Governance (Osborne, 2010). This framework does not replace

1_Good Public Governance

Public governance strategy

P2p governance-public-services

INTERNATIONAL FRAMEWORK: GOOD GOVERNANCE IN THE PUBLIC SECTOR › system › ...Good-Governance-in-the-Public-Sector-IFA… · Good governance in the public sector encourages better

TechnologicalForecasting&SocialChange ...mosb/public/pdf/2284/Frey and Osborne - 201… · C. Frey, M. Osborne / Technological Forecasting & Social Change 114 (2017) 254–280 255

Gwinnett County Public Schools Parent Portal Overview Osborne Middle School